GIFT OF 
 MICHAEL REESE 
 
ASSAYING 
 
BY 
 
 SrOTTISWOODE AXD CO., NEW-STREET SQUARE 
 LONDON" 
 
A MANUAL 
 
 OF 
 
 PEACTICAL ASSAYING 
 
 BY 
 
 JOHN MITCHELL, F.C.S. 
 
 EDITED BY 
 
 WILLIAM CBOOKES, F.E.S., PRES. C.S. 
 
 SIXTH EDITION 
 
 ILLUSTRATED WITH 201 WOODCUTS 
 
 LONDON 
 LONGMANS, GBEEN, AND CO. 
 
 1888 
 
 All rights reserved 
 

PBEFACE 
 
 TO 
 
 THE SIXTH EDITION. 
 
 THE last edition of Mitchell's Assaying was published in 
 1881, and a new edition being required, the opportunity 
 has been taken of x rewriting some of the descriptions 
 which the progress of research had made old or ill adapted 
 to modern requirements. At the same time much new 
 matter has been introduced, and matter which had become 
 obsolete has been omitted. The number of woodcuts has 
 been increased from 188 in the last edition to 201 in the 
 present edition, whilst the number of pages, notwithstand- 
 ing numerous excisions, has grown from 809 to 896. 
 
 Among the new and important additions to this edition 
 may be enumerated a description of the ' Automatic 
 Sampling Machine,' invented by Mr. D. W. Brunton ; 
 many new gas furnaces and burners for the laboratory, 
 devised by Mr. Fletcher, Messrs. J. J. Griffin, and others ; 
 new blowpipe reagents and operations ; new processes, 
 dosimetric, volumetric, and colorimetric, for the partial 
 and complete assay of iron ores, iron, steel, spiegeleisen, &c. 
 In the copper assay a full description is given, for the 
 first time in this country, of the American system of fire 
 
[vi] PREFACE TO THE SIXTH EDITION. 
 
 assay. The system adopted at Swansea is so interwoven 
 with the customs of the trade that its replacement by a 
 more accurate process is perhaps not to be expected for 
 some time, but for those assay ers who are not bound to 
 this particular process the Lake Superior system can be 
 strongly recommended, as being quick, inexpensive, and 
 comparing favourably in accuracy with the wet methods. 
 
 In the assay of silver the action of bismuth on the 
 ductility of this metal a subject hitherto overlooked 
 has received considerable attention. Much has also been 
 added on the subject of gold ores, a matter of large 
 and increasing interest to miners and metallurgists ; and 
 improved modes of assaying the precious metal and its 
 detection in poor ores are given. Besides these more 
 important additions and alterations, minor additions are 
 to be found in every chapter. 
 
 A little more prominence has been given to the English 
 system of grain weights, as English assayers are more 
 familiar with these than with the metric system in use on 
 the Continent ; and, as far as possible, the decimal system 
 has been adopted, the grain being taken as the unit. So 
 many valuable memoirs on assaying and metallurgical 
 subjects are published abroad that it has, however, been 
 found impracticable to adhere universally to grains. 
 
 W. C. 
 
 LONDON : April 1888. 
 
EXTRACTS FROM THE 
 
 PEEFACE TO THE THIED EDITION. 
 
 IN this edition are incorporated all the late important 
 discoveries in Assaying made in this country and abroad, 
 and special care is devoted to the very important Volumetric 
 and Colorimetric Assays, as well as to the Blowpipe Assays. 
 Most of the chapters are entirely rewritten, whilst the 
 chapter on Crystallography being a subject only remotely 
 bearing on Assaying is left out altogether. On the 
 other hand, in some cases it may seem that by treating 
 of purely analytical details the limits -of Assaying have 
 been exceeded. But these departments are so closely 
 related as to make it impossible to fix the line of demar- 
 cation between them. Moreover, chemistry is cultivated 
 by almost all to whom this work is of interest or service, 
 so that it is hoped these amplifications will add to its value. 
 The old equivalents are retained, as they are more generally 
 understood by students of science who do not make 
 chemistry their chief study. 
 
 The Editor is under many obligations to his friend Dr. 
 Eohrig, M.E., for assistance in revising the manuscript and 
 incorporating into the work the latest continental improve- 
 ments, as set forth in Professor Kerl's Probirkunst. The 
 author of the best work on volumetric analysis which has 
 
[viii] EXTKACTS FROM PEEFACE TO THE THIRD EDITION. 
 
 yet appeared in English, Mr. Sutton, F.C.S., has kindly 
 placed several cuts, &c., at the Editor's disposal, and some 
 descriptions of German processes have been taken from the 
 last English Edition of Fresenius's Quantitative Analysis a 
 work which should be a standard of reference for all 
 students who desire to carry their chemical researches 
 further than is possible to be treated of in a work pro- 
 fessing to deal only with Assaying. 
 
 LONDON : September 1868 
 
PBEFACE 
 
 TO 
 
 THE FIEST EDITION. 
 
 WHEN the rank our country holds among nations, as 
 regards her Mining interest, is taken into consideration, it 
 must be with all a matter of surprise that no work especially 
 devoted to the elucidation of the processes to be employed 
 in ascertaining the richness in rnetal of any sample of ore 
 (that is, in other terms, its Assay) has of late years appeared 
 before the British public. Indeed, the only work at pre- 
 sent known in England is Berthier's ' Traite des Essais par 
 la Voie Seche,' which, for the mere purpose of inculcating 
 the principles of Assaying, has many disadvantages not 
 the least of which is its being written in a foreign tongue ; 
 and although a knowledge of French is now so very 
 general, yet many are prevented buying scientific works 
 in that language on account of the difficulties of finding 
 equivalents for the technicalities which must necessarily be 
 employed. It is also a very large work, and one contain- 
 ing much matter which the assayer does not need matter 
 relating to the composition of wood and coal ashes, furnace 
 products, &c., which are more especially adapted for the 
 metallurgist. 
 
[x] PREFACE TO 
 
 These considerations, coupled with the paucity of any 
 knowledge of Assaying, excepting that confined to a very 
 limited number of persons, induced the author of the 
 following pages to turn a considerable amount of his at- 
 tention to this subject, more especially as much difficulty 
 was experienced in not having a suitable text book for the 
 use of his pupils. A portion of the following pages was 
 drawn up as a Manual for such a purpose ; but, on con- 
 sideration, it was thought the extension of such a work 
 was so much needed that it was determined to alter the 
 original plan as> far as was consistent with the complete 
 carrying out of the object in view, viz. the production of 
 a Manual embodying information in every branch of 
 Assaying, either by the wet or the dry processes. 
 
 The following is a sketch of the manner in which this 
 is accomplished, the author having followed the excellent 
 arrangement of Berthier as closely as possible, from whose 
 work also much matter that suited these pages, and which 
 it would have been useless to rewrite, has been inserted. 
 First, the Mechanical and Chemical Operations of Assay- 
 ing are treated in full, inclusive of a description of the 
 apparatus required, their mode of use, &c. Secondly, 
 Furnaces, Fuel, and Crucibles, together with a description 
 of the best Pyrometers, and their applications. Thirdly, 
 the Fluxes, their properties, preparation, use, &c. Fourthly, 
 an Essay on the use of the Blowpipe, and all its appurten- 
 ances ; as Fluxes, Supports, &c. Fifthly, the action of 
 the Fluxes on some Mineral Substances. Sixthly, a method 
 of discriminating many Minerals by means of the Blow- 
 pipe, aided by a few tests by the humid method. Seventhly, 
 the Humid Analysis of many Mineral Substances, their 
 composition, locality, &c. (All the minerals mentioned in 
 the three last heads comprehend such only as generally 
 
THE FIRST EDITION. [xi] 
 
 came under the notice of the Assay er.) Eighthly, the 
 complete Assay of all the common Metals, in addition to 
 which the Assay of Sulphur, Chromium, Arsenic, Heating 
 power of Fuel, &c., is fully discussed ; and ninthly, and 
 lastly, a copious Table drawn up for the purpose of ascer- 
 taining in Assays of Gold and Silver the precise amount, 
 in ounces, pennyweights, and grains, of Noble metal con- 
 tained in a Ton of Ore from the assay of a given quantity 
 This Table is the most complete and copious yet published 
 
 Not only has it been endeavoured to collect all that is 
 generally known on the subject of Assaying, but many 
 new facts have been added, and such matter entered into, 
 that the success of an assay is rendered much more certain ; 
 and most assays are conducted more rapidly and with 
 greater exactitude than heretofore. 
 
 It has also been endeavoured to introduce a new system, 
 in which is pointed out the rationale of each process, with 
 the chemical action taking place between the fluxes and 
 the ores in course of assay, so that by paying a careful 
 attention to the matters discussed^ so much of the chemical 
 nature of all ores that can come under the assay er's hand 
 may be known, that the practice by * rule of thumb ' (a 
 rule on which very little dependence is to be placed, ex- 
 cepting after years of the most laborious practice, and a 
 rule which cannot be imparted, excepting the pupil pursue 
 the same unprofitable course) must, it is hoped, be speedily 
 abandoned when, by knowing the chemical properties of the 
 body operated on, the necessary fluxes and processes might 
 be at once indicated, and with a certainty of perfect success. 
 
 Having premised thus much, the author must beg to 
 express his thanks to his friend Mr. F. Field for the kind 
 
[XllJ PREFACE TO THE FIRST EDITIOX. 
 
 assistance lie afforded him whilst experimenting on the 
 various modes of assay described in the body of the work ; 
 and trusting that any little imperfections which may be 
 detected will not be harshly criticised, but that it may be 
 taken into consideration that the author has attempted to 
 improve a branch of mining knowledge to which unfor- 
 tunately too little attention has been devoted, and to which, 
 if he has added anything useful, he is indebted for the first 
 principles of such knowledge to Berthier's ' Traite des 
 Essais,' for which, to the talented writer of the above 
 work, he is under the most lasting obligation. 
 
CONTENTS. 
 
 CHAPTER I. 
 
 PAGE 
 
 Chemical nomenclature Laws of combination, etc. . . 1 
 
 Metallic and non-metallic elements .... 2 
 
 Oxides ........ 4 
 
 Salts ....... .. 4 
 
 Binary compounds containing no oxygen . ... 5 
 
 Laws of combination ...... 5 
 
 Chemical symbols ; their employment and uses . . ,7 
 
 CHAPTER II. 
 
 Preparation of the sample. Weighing . ... 9 
 
 Automatic sampling machine .' . . . .11 
 
 Anvil and stand, 14. Hammers, 15. Cold chisel, 16. 
 
 Shears, 17. 
 Pestle and mortar . . . . . . .17 
 
 Iron mortar, 17. Wedgwood mortar, 17. Porcelain mortar, 
 
 18. Steel mortar, 19. Agate mortar, 20. 
 The sieve ........ 21 
 
 Elutriation . . . . . . .22 
 
 Washing, dressing, or vanning . . . . .23 
 
 The balance . . . . . . .26 
 
 Operation of Aveighing, 26. Bullion balance, 27. Rough 
 assay balance, 28. Assay balance, 28. Conditions of accu- 
 racy and delicacy, 29. 
 The weights . . . . . . .34 
 
 Assay weights for silver, 35. Assay weights for gold, 35. 
 Method of weighing . . . . .36 
 
 Mayer's method, 39. 
 Incinerating precipitates before weighing . . . .41 
 
[xiv] CONTENTS. 
 
 CHAPTER III. 
 
 PAGE 
 
 General preparatory chemical operations . . . .42 
 
 Calcination . . . . . . .42 
 
 Use of Crucibles, 42. 
 
 Roasting . . . . . . . .44 
 
 Roasting test, 44. Roasting in crucibles, 45. Roasting in 
 
 platinum capsules, 46. 
 
 Reduction ...... , . 46 
 
 Cementation, 47. Reduction by hydrogen 48. 
 Fusion ........ 48 
 
 Solution ........ 49 
 
 Glass and platinum forceps, 50. 
 
 Distillation . . . . . . .51 
 
 Liquid distillation, 51. Dry distillation, 53. Sublimation, 
 
 54. 
 
 Scorification Cupellation. . . . . .54 
 
 CHAPTER IY. 
 
 Production and application of heat . . . .55 
 
 Calcining furnace . . . . . . .55 
 
 Chimney . . . . . . . .56 
 
 Evaporating furnaces . . . . . .57 
 
 The hood ........ 57 
 
 Fusion furnaces. Wind furnaces. . . . .57 
 
 Body of furnace, 58. Ashpit, 57. Bars, 58. Chimney, 59. 
 Blast furnaces . . . . . . .60 
 
 Royal Institution furnace, 60. Sefstrom's blast furnace, 61. 
 
 Deville's furnace, 62. 
 Muffle or cupel furnace . . . . . .62 
 
 Aufrey and d'Arcet's furnar ^0. Brick furnace for numbers 
 
 of cupellations, 64. Ur .sal furnace, 65. 
 Furnace operations . . . . .67 
 
 Auxiliary a^^aratus . . . . . .67 
 
 Poker- or stirring rods, 67. Tongs, 67. Spectacles and look- 
 ing glass, 68. Curved rods, 68. "Wrought iron ladle, 69. 
 Ingot moulds, 69. 
 Fuel for furnaces . . . ... . .69 
 
 Coke, 70. Charcoal, 70. 
 
 Effects produced by wind and blast furnaces . . .72 
 
 Oil and gas blast furnaces . . . . . .74 
 
 Oil furnaces . . . . . . .74 
 
 Description of the apparatus, 74. Oil lamp furnace, 75. 
 Oil reservoir, 75. Crucible furnace, 75. Management of 
 the oil lamp furnace, 76. Power of the oil lamp furnace, 
 79. Requisite blowing power, 79. 
 
CONTENTS. [xv] 
 
 PAGE 
 
 Griffin's gas furnace t . . . . .80 
 
 Operations in crucibles, 80. Operations in muffles, 89. 
 Skittle-pots, 83. Flue, 83. Distillation of zinc, 84. Extra 
 large gas furnace, 84. 
 Fletcher's universal gas furnace . . . . ,84 
 
 Burner, 84. Single jacketed arrangement, 84. Double 
 
 jacketed chemical furnace, 84. 
 Brown's gas assay furnace . , . . . .87 
 
 Fletcher's reverberatory gas furnace . . . .90 
 
 Fletcher's muffle and draught crucible furnaces . . .92 
 
 Fletcher's new melting arrangement . . . .92 
 
 Injector gas furnace, 92. The burner, 93. ' Salamander ' 
 crucibles, 93. Injector gas or spirit furnace, 94. Benzo- 
 line, 95. 
 Gore's gas furnace . . . . . . .95 
 
 Griffin's reverberatory gas furnace . , .100 
 
 Melting furnace for lead, tin, antimony, etc. . . .104 
 
 Bunsen's gas burner, 105. Mounts for crucibles, 108. Gas 
 furnace for boiling or evaporating, 110. Solid flame gas 
 burner, 111. Dirt proof high power burner, 112. Safety 
 Bunsen burner, 113. 
 Lutes and cements . . . . . .113 
 
 Fire lute, 113. Fat lute, 113. Roman cement, 114. Plas- 
 ter of Paris, 114. Linseed or almond meal, 114. Lime 
 and egg lute, 114. White lead with oil, 114. Yellow 
 wax, 114. Soft cement, 114. Cement for brass or glass, 
 
 114. Cement for mending pestles, 115. Adhesive paste, 
 
 115. Waterproof cement, 116. Resinous or hard cement, 
 117. Caoutchouc, 117. Faraday's directions for luting 
 iron, glass, or earthenware retorts, tubes, etc., 118, 
 Willis's cement, 119. Iron cemep* , -\J 9. Beal's cement, 
 
 120. Boiler cement, 120. Bruyei rement, 120. Oxy- 
 chloride of zinc cement, 1 20. 
 
 Crucibles, cupels, etc. . . . . . V ' 1 20 
 
 Hessian, Cornish, Stourbridge, and London clay crucibles, 1 20. 
 ' Salamander ' brand of crucibles and plumbago fittings, 
 
 121. Porcelain crucibles, 122. Black-lead crucibles, 123. 
 Charcoal crucibles, 126. Lined crucibles, 127. Lime 
 crucibles, 128. Forbes's experiments with lime and char- 
 coal crucibles, 129. Alumina crucibles, 130. Magnesia 
 crucibles and bricks, 130. Malleable iron crucibles, 135. 
 Platinum crucibles, 135. Preserving platinum crucibles, 
 135, 137, 138. Dexter's lamp for platinum crucibles, 135. 
 Silver crucibles, 140. Nickel crucibles, 140. 
 
 Cupels . . 141 
 
 Cupel mould, 142. Powder for cupels, 143. Scorifiers, 144. 
 Methods of measuring the heat of furnaces . . .144 
 
[xvi] CONTENTS. 
 
 PAGE 
 
 Wedgwood's pyrometer, 144. Daniel's pyrometer, 144. 
 Wilson's pyrometer, 147. Siemens's pyrometer, 148. 
 Comparing the temperatures of two furnaces, 149. 
 
 CHAPTER Y. 
 
 Fuel, its assay and analysis . . . . .150 
 
 External appearance of fuel, its porosity, compactness, fracture, 
 
 size, and shape of pieces . . . . .152 
 
 Estimation of adhering water . . . . .152 
 
 Estimation of specific gravity . . . . .153 
 
 Estimation of absolute heating power . . . .154 
 
 Pyrometric heating power, 154. Different modes of ascer- 
 taining absolute heating power, 154. Berthier's method, 
 " 155. ' lire's method, 157. lire's calorimeter, 158. 
 Wright's calorimeter, 159. 
 Estimation of specific heating power . . . .160 
 
 Estimation of pyrometric heating power . . . ..160 
 
 Estimation of the volatile products of carbonisation . .160 
 
 Examination of the coke or charcoal left behind on carbonisation 161 
 Estimation of the amount of ash . . . . .162 
 
 Estimation of the amount of sulphur . . . .162 
 
 Examination of other peculiarities of fuel . . .164 
 
 Calculation of results . . . . . .164 
 
 Assay of coal before the blowpipe . . . . .165 
 
 Valuation of coal for production of illuminating gas - . .167 
 
 CHAPTER VI. 
 
 Reducing agents . . . . . . .169 
 
 Hydrogen gas, 169. Carbon, 170. Black-lead or graphite, 
 170. Estimation of the value of graphite, 170. Anthra- 
 cite, 171. Coke, 171. Wood charcoal, 171. The fatty 
 oils, 172. Tallow, 172. Resins, 172. Sugar, 173. 
 Starch, 173. Gum, 174. Tartaric acid, 174. Oxalic 
 acid, 174. Ammonium oxalate, 174. Comparative reduc- 
 ing powers of the above agents, 175. Metallic iron, 175. 
 Metallic lead, 176. 
 
 Oxidising agents . . . . . . .176 
 
 Litharge, 176. Ceruse or white lead, 176. Lead silicates 
 and borates, 177. Potassium and sodium nitrates. 177. 
 Assay of saltpetre, 178. Lead nitrate, 180. Manganese 
 peroxide, 181. Copper oxide, 181. Iron peroxide, 181, 
 The caustic alkalies, potash and soda, 181. Potassium 
 and sodium carbonates, 181. Lead, copper, and iron sul- 
 phates, 181. Sodium sulphate, 181. 
 
CONTENTS, [xvii] 
 
 PAGE 
 
 Desulphurising agents . . . . . .181 
 
 Atmospheric oxygen, 182. Charcoal, 182. Iron, 182. 
 
 Litharge, 182. Behaviour of litharge with sulphides of 
 
 manganese, 183 ; iron, 183 ; copper, 185 ; antimony, 186 ; 
 
 zinc, 186 ; lead, 187. Caustic alkalies and their carbonates, 
 
 187. Nitre, 188. Lead nitrate, 188. Lead sulphate, 
 
 188. 
 Sulphurising agents . . . . . .189 
 
 Sulphur, 189. Cinnabar, 189. Galena, 189. Antimony 
 
 sulphide, 189. Iron pyrites, 190. Alkaline persulphides, 
 
 190. 
 
 Fluxes . 191 
 
 Non-metallic fluxes . . . . . .191 
 
 Silica, 192. Lime, magnesia, alumina, and their silicates, 192. 
 
 Alumina, 193. Borax, 193. Glass, 193. Analysis of 
 ' different kinds of glass, 194. Fluor-spar, 195. Potassium 
 
 and sodium carbonates, 195. Potassium nitrate, 196. 
 
 Common salt, 196. White flux, black flux, and raw flux, 
 
 197. Argol or cream of tartar, 199. Salt of sorrel, or 
 
 potassium binoxalate, 199. White and mottled soap, 199. 
 
 Reducing power of the various fluxes, 200. 
 Metallic fluxes . . . . . . .201 
 
 Litharge and ceruse, 201. Glass of lead (lead silicate), 201. 
 
 Lead borate, 201. Lead sulphate, 201. Copper oxide, 
 
 201. The iron oxides, 201. 
 
 CHAPTER VII. 
 
 The blowpipe and its uses ...... 202 
 
 Blowpipe, 203. Trumpet mouthpiece, 204. Nipples, 205. 
 Lamps and oil, 206. Faraday's directions for using the 
 blowpipe, 206. Oxidation, 209. Reduction, 209. 
 
 Auxiliary blowpipe apparatus . . . . .210 
 
 Supports, 210. Charcoal, 210. Platinum, 211. Wire, 211. 
 Forbes's instruments for preparing charcoal, 211. Plati- 
 num spoon, 212. Aluminium support, 212. Asbestos 
 cardboard, 213. 
 
 Reagents and fluxes . . . . . .213 
 
 Blue litmus paper, 213. Reddened litmus paper, 213. 
 Brazil-wood paper, 213. Turmeric paper, 213. Nitric 
 acid, 213. Zinc, 213. Copper, 213. Iron wire, 213. 
 Potassium cyanide, 214. Sodium carbonate, 214. Re- 
 duction of metallic oxides, 216. Borax, 218. Sodium 
 ammonio-phosphate, 219. Potassium nitrate, 220. Potas- 
 sium bisulphate, 220. Sodium bisulphate, 221. Vitrified 
 boracic acid, 221. Cobalt nitrate, 221. Nickel oxalate, 
 
 a 
 
[xviii] CONTENTS. 
 
 PAGE 
 
 221. Copper oxide, 221. Silica, 222. Turner's flux, 222. 
 Ammonium fluoride, 222. Calcium fluoride and sulphate, 
 223. Bone ashes, 224. Proof lead, 224. Tinfoil, 224. 
 Dry silver chloride, 224. Tincture of iodine, 229. Dry 
 silver iodide, 231. Soda-paper, 233. Forbes's soda-paper 
 apparatus, 233. 
 
 General routine of blowpipe operations . . . .234 
 
 Size of the assay, 234. Order of blowpipe operations, 234. 
 
 Discrimination of minerals . . . . .237 
 
 Crystalline form, 237. Mode of fracture, 238. Lustre, 239. 
 Colour and streak, 239. Hardness, 240. Specific gravity, 
 241. Sonstadt's solution, 242. Fusibility, 244. Chemical, 
 characters, 244. Colour of borax bead, 246. Appearance 
 of reduced bead, 247. Requirements for testing minerals, 
 248. 
 
 Description of minerals . . . . . .248 
 
 Quartz and the silicates, 248. Rock crystal, 248. Amethyst, 
 248. Rose quartz, 250. Cairngorm, 250. False topaz, 
 250. Chalcedony, 250. Carnelianand sard, 250. Agate, 250. 
 Onyx or sardonyx, 250. Flint or hornstone, 250. Jasper, 
 250. Bloodstone, 250. Opal, 250. Talc, 250. Mica, 251. 
 Chlorite, 251. Serpentine, 251. Meerschaum and ne- 
 phrite, 251. Augite and horneblende, 252. Chrysolite, or 
 olivine, 252. Tourmaline, 252. Garnet, 253. Topaz, 
 253. Beryl or emerald, 254. Zircon or hyacinth, 254. 
 Felspar, 255. Mica, 255. Zeolites, 255. Corundum or 
 sapphire, 256. Spinel, 256. Chrysoberyl, 257. Diamond, 
 
 257. Graphite or black-lead, 258. Coal, 258. Apatite, 
 
 258. Fluor-spar, 259. Calcspar, 259. Magnesite, 260. 
 Dolomite, 260. Aragonite, 260. Rock salt, 261. Solu- 
 ble sulphates, 261. Nitre, 261. Gypsum, selenite, ala- 
 baster, 261. Heavy spar or barytes, 262. Sulphur, 262. 
 Tin ore, 262. Molybdenite, 263. Bismuth, 263. Anti- 
 mony sulphide, 263. Arsenic, 264. Arsenic sulphide, 
 264. Native iron, 265. Iron pyrites, 265. Arsenical 
 pyrites, 265. Magnetic iron, 266. Specular iron, hema- 
 tite, or micaceous iron, 266. Red ferric oxide or red 
 ochre, 266. Brown ferric oxide, or brown ochre, 267. 
 Titanic iron, 267. Chrome iron, 267. Green-earth, 267. 
 Iron carbonate, 268. Manganese ores, 268. Arsenical 
 nickel, 268. Smaltine or tin-white cobalt, 269. Cobalt 
 bloom, 269. Blende, 269. Zinc carbonate, 270. Zinc 
 silicate, 270. Galena, 270. Lead carbonate or cerusite, 
 270. Pyromorphite, 271. Lead sulphate, 271. Cinnabar, 
 272. Native mercury or quicksilver, 272. Native 
 copper, 273. Vitreous copper, 273. Copper pyrites, 273. 
 Grey copper, 274. Black cupric oxide, 274. Red 
 
CONTENTS. [xix] 
 
 PAGE 
 
 cuprous oxide, 274. Copper carbonates, blue and green, 
 275. Platinum, 275. Gold, 275. Silver, 276. Silver 
 sulphide, 276. Antimonial and arsenical silver ores, 276. 
 Horn silver, 277. Lead and antimony sulphides, 277. 
 Mercury, lead, silver, or copper selenides, 277. Millerite 
 or nickel sulphide, 277. White nickel, 278. Rutile, 278. 
 Sphene, 278. Wolfram, 278. Pitchblende, 278. Mag- 
 netic iron pyrites, 279. 
 
 Determination of minerals . . . . .279 
 
 Scheme for the determination of minerals .... 280 
 
 Group I. Minerals which have a metallic lustre, and which give 
 
 off sulphur . . . . . . .281 
 
 Group II. Minerals which have a metallic lustre, and which give 
 off either an odour of garlic without sulphur, or white 
 fumes which have not a garlic or sulphurous odour . 283 
 
 Group III. Minerals which have a metallic lustre, and which 
 
 give off no fumes ...... 284 
 
 Group IV. Minerals which possess a non-metallic lustre, a 
 coloured streak, and which give off fumes or odour when 
 heated before the blowpipe .... 286 
 
 Group V. Minerals which possess a non-metallic lustre, a 
 coloured streak, but which give off no fumes or odour 
 before the blowpipe . . . . .287 
 
 Group VI. Minerals which have a non-metallic lustre, and which 
 
 are scratched by quartz, showing a white streak . . 288 
 
 Group VII. Minerals which have a non-metallic lustre, and which 
 
 are not scratched by quartz . . . .291 
 
 Coloured flames . . . . . . .294 
 
 Blue flames, 294. Green flames, 294. Yellow flames, 295. 
 Red flames, 295. Chlorine, 295. Lead, 295. Arsenic, 
 
 295. Selenium and antimony, 295. Bromine, 295. Boracic 
 acid, 295. Tellurium, 296. Copper, 296. Iodine and 
 copper, 296. Phosphoric acid, 296. Baryta, 296. Zinc, 
 
 296. Soda, 296. Water, 296. Strontia, 296. Lithia, 
 296. Lime, 297. Potash, 297. 
 
 CHAPTER VIII. 
 
 Volumetric analysis . . . . . . .298 
 
 Reactions of volumetric analysis, 298. Principle of volume- 
 tric analysis, 298. Standard solutions, 302. 
 Instruments and apparatus, 302. The burette, 303. Mohr's 
 burette, 304. Gay-Lussac's burette, 304. Britton's burette, 
 305. 
 
 Modes of operating, 305. The pipette, 306. Measuring 
 flasks, 307. 
 
 a 2 
 
[xx] CONTENTS. 
 
 CHAPTER IX. 
 
 PAGE 
 
 The assay of iron ....... 308 
 
 Ores of iron ....... 308 
 
 Magnetic iron ore, 308. Red hematite, 308. Brown hema- 
 tite, 308. Spathic iron ore, 308. Titaniferous iron ore, 
 308. Franklinite, 308. Clay band and black band iron 
 stone, 308. 
 Assay of iron in the dry way ..... 309 
 
 Classification of iron ores, 309. Fluxes, 309. Air furnaces, 311. 
 Crucibles, 311. The charge, 313. Ores of unknown com- 
 position, 313. Ores previously analysed, 314. The opera- 
 tion, 315. 
 Effects produced by the following substances : 
 
 Manganese, 317. Titanium, 317. Phosphorus, 317. Sul- 
 phur, 317. Chromium, 317. 
 Calculation of results . . . . . .319 
 
 Assay of iron and its ores in the wet way . . .321 
 
 Dr. Penny's process . . . . . .321 
 
 Reduction by zinc, 327. Reduction by stannous chloride, 328. 
 
 Reduction by ammonium bisulphite, 328. 
 Titration of iron with sodium hyposulphite, 330. Oudemans's 
 
 method, 330. Haswell's method, 330. 
 The complete assay of iron ores ..... 332 
 
 The mechanical treatment, 332. The chemical treatment, 
 334. Estimation of phosphoric acid, 334. Sulphur and 
 iron, 338. Silica, ferric oxide, alumina, manganese, lime, 
 and magnesia, 341. Nickel, cobalt, and zinc, 344. Esti- 
 mation of ferrous oxide, 346. Ferrous oxide in insoluble 
 silicious matter, 347. Alumina, 349. Calculation of the 
 analysis, 349. Carbonic acid, 349. Water and carbon in 
 carbonaceous matter, 351. Alkalies, 355. Copper, lead, 
 arsenic, and antimony, 356 Titanic acid, 358. Estima- 
 tion of specific gravity, 360. 
 Gooch's perforated crucible . . . . .361 
 
 Estimation of carbon, silicon, phosphorus, &c., in metallic iron and 
 steel, 363. Boring and sampling, 363. Estimation of total 
 carbon, 367. Weyl's method, 368. Estimation of gra- 
 phite, 369. Dr. Eggertz's method, 369. Mr. Tosh's 
 method, 370. Estimation of combined carbon, 371. O. 
 Arnold's method, 373. J. Blodgett Britton's method, 374. 
 Mr. Stead's method of estimating minute quantities of 
 carbon, 376. A new form of chromometer, 380. 
 Preparation of inorganic standards for the colorimetric carbon 
 
 test . . . . . . . .381 
 
 Day standard colours, 382. Night standard colours, 386. 
 Estimation of sulphur in iron and steel .... 388 
 
CONTENTS. [xxi] 
 
 PAGE 
 
 Dr. Eggertz's method, 388. J. Wiborgh's method, 393. 
 
 Preparation of the cloth, 394. 
 The colour scale, 398. Details of the process, 401. Estimation 
 
 of sulphur in iron ores, 405. 
 Estimation of silicon in iron and steel . . . .407 
 
 Mr. Turner's method . . . . . ,413 
 
 Estimation of basic cinder and oxides in manufactured iron . 414 
 
 Mr. Bettel's process . . . . . .415 
 
 Estimation of phosphorus in iron and steel , . .416 
 
 M. Tantin's method, 416. J. B. Mackintosh's method, 417. 
 J. Lawrence Smith's method, 417. Quantity of iron em- 
 ployed, 419. Solution, 420. Concentration of the phos- 
 phorus, 420. Separation of the phosphorus, 421. 
 Estimation of manganese in iron . . . . .422 
 
 E. Riley's method, 424. The direct method, 424. The in- 
 direct method, 425. W. Kalman and A. Smolka's method, 
 426. Dr. Peter's method, 427. Mr. Galbraith's method, 
 429. 
 Estimation of titanium in iron . ... . . 430 
 
 Mr. Riley's method, 430. Mr. Bettell's method, 432. 
 Turner's table of the hardness of iron and steel 433 
 
 CHAPTER X. 
 
 The assay of copper ....... 434 
 
 Classification of minerals and substances containing copper . 434 
 
 Class I. Sulphuretted ores or products with or without sele- 
 nium, antimony, or arsenic, 434. Copper glance, 434. 
 Chalcopyrite, 434. Erubescite, 434. Bournonite, 434. 
 Fahlerz, 434. Covelline, 434. Wolfsbergite, 434. Do- 
 meykite, 434. Copper regulus, copper speiss, etc., 434. 
 
 Class II. Oxidised ores and products, 434. Red copper, 434. 
 Malachite, 434. Azurite, 434. Cyanosite, 434. Phos- 
 phate of copper, 434. Arseniates, 434. Chromate, vana- 
 date, and silicate of copper, slags, etc., 434. 
 
 Class III. Copper and its alloys, 434. 
 Classification of different methods of assaying copper . . 435 
 
 A. Assay in the dry way, 435. 
 
 a. For rich ores and products of Class I., 435. 
 English copper assay ...... 435 
 
 Moissenet's description, 435. Ticketing in Cornwall, 435. 
 De la Beche's sketch of the method, 436. Division 
 adopted, 436. 
 
 Section I. Reactions, 437. Two kinds of assays roasted and 
 raw sample, 437. 
 
[xxii] 
 
 CONTENTS. 
 
 PAGE. 
 
 1. Regulus, 438. Pyrites, 438. Very poor pyrites, 438. 
 
 Variegated copper ore, 439. Sulphide of copper, 439. 
 Carbonated minerals, 439. Native mixture, 439. 
 
 2. Calcining, 440. 
 
 3. Coarse copper, 440. 
 
 4. Washings, 441. 
 
 5. Testing, refinery, 442. 
 
 6. Slags for prill, 443. 
 
 Section II. Manipulations, 443. Crucibles used in Cornwall, 
 444. Wind furnace, 444. Fusion for regulus, 445. Cal- 
 cining the matt, 448. Coarse copper, 449. Washings^ 
 
 450. Testing and refining, 450. Prill, 450. 
 
 Section III. Some minerals and substances of a special nature ; 
 influence of foreign metals, 451. Stanniferous minerals, 
 
 451. Antimonial minerals, 451. Zinciferous minerals, 452. 
 Plumbiferous minerals, 452. Regulus of Chili, 452. Slags 
 of copper, 452. Old copper, 452. 
 
 Section IY. Summary considerations comparison of the 
 results with the analysis by the wet way, 453. Principal 
 causes of loss in the Cornish method, 454. Assay of 
 copper in the dry way, 454. 
 b. For ores and products of Class II., 455. 
 Lake Superior fire assay ...... 455 
 
 Dr. Peter's description, 455. Sampling, 456. Fluxes, 457. 
 Furnace, 457. 
 
 B. Assays in the wet way, 461. 
 
 a. Colorimetric copper assay, 461. 
 
 Heine's method, 461. Le Play's method, 46 7. T.O. Cloud's 
 method, 467. Endemann's method, 467. Carnelly's 
 method, 470. Standard copper solution, 470. Solution 
 of, 472. Jacquelain's and Von Hubert's colorimetric 
 assay, 472. 
 
 b. Volumetric copper assays, 476. 
 
 Fleck's modification of Mohr's method, 476. E. 0. Brown's 
 method by sodium hyposulphite, 479. 
 
 c. Electrolytic copper assay, 480. 
 
 Estimation of copper in the Mansfield ores by Dr. Steinbeck's 
 
 process ....... 480^ 
 
 Extraction of the copper from its ores, 480. Separation of 
 the copper, 481. Quantitative estimation of the precipi- 
 tated copper, 482. Special observations on this method, 
 483. 
 Estimation of copper in the Mansfield ores by M. C. Luckow's 
 
 process . . . . . . : , 487 
 
 Roasting the ore, 489. Solution of the roasted product, 489. 
 Precipitation of the copper, 490. Weighing the copper, 493. 
 Assay of copper pyrites . . . . . .494 
 
CONTENTS. [xxiii] 
 
 PAGE 
 
 Detection of traces of copper in iron pyrites and other bodies . 497 
 Estimation of arsenic in copper . . . . .499 
 
 CHAPTER XI. 
 
 The assay of lead . . . . . ,502 
 
 Classification of minerals and substances containing lead . . 502 
 
 Action of reagents on lead sulphide, 502. Oxygen, 502. Me- 
 tallic iron, 502. Alkalies and alkaline carbonates, 503. 
 Potassium nitrate, 503. Argol, 503. 
 
 Assay of substances of the first class (sulphides, antimonides, etc.) 504 
 Processes generally adopted : 
 
 1. Fusion with potassium carbonate, 504. 
 
 2. Fusion with black flux, 510. 
 
 3. Fusion with metallic iron, 511. 
 
 4. Fusion with sodium carbonate, or black flux and metallic 
 
 iron, 514. 
 
 5. Roasting and reducing assay, 515. Assay with black 
 
 flux and iron, 517. Roasting and reduction assay 
 with iron, 518. Roasting and fusing with black flux, 
 518. Level's fusion assay with potassium ferrocyanide 
 and cyanide, 519. 
 
 6. Assay with sulphuric acid, 519. 
 
 7. Assay of galena in the wet way, 521. 
 
 Assay of substances of the second class . . . .523 
 
 Humid assay of ores of the second class . . . .526 
 
 Assay of substances of the third class . . . .527 
 
 Humid assay of substances of the third class . . . 528 
 
 Maxwell Lyte's process, 529. Mascazzini's process, 529. 
 
 Jannesay's process, 529. 
 Alloys of lead (Class IV.) 530 
 
 Assay with sulphuric acid, 530. 
 
 Level's fusion assay with potassium ferrocyanide and cyanide . 531 
 Estimation of lead by means of standard solutions . . 531 
 
 1. Flores Dumonte's method, 531. 
 
 2. Schwartz's method, 532. 
 
 3. Buisson's volumetric process, 533. 
 
 4. Diehl's volumetric process, 534. 
 
 CHAPTER XII. 
 
 Assay of tin ..... 537 
 
 Tin ores ...... 537 
 
 Tin oxide, 537. Crystallised tin oxide, 537. Disseminated 
 tin oxide, 537. Sandy tin oxide, 537. Concretionary tin 
 oxide, wood tin, 537. 
 
[xxiv] CONTENTS. 
 
 PAGE 
 
 Analysis of a sample of tin oxide from Cornwall, 538. 
 Remarks on tin ore and the minerals which may be mistaken 
 
 for it, by Dr. A. Leibius, 538. 
 Assay of pure tin oxide ...... 539 
 
 Method used in Cornwall, 540. Method by means of potas- 
 sium cyanide, 540. 
 Assay of tin oxide mixed with silica .... 542 
 
 Assay of tin ores containing silica and tin slags . . .542 
 
 Assay of tin ores containing arsenic, sulphur, and tungsten . 543 
 
 Approximative assay ...... 544 
 
 J. H. Talbot's method for assaying tin in the presence of 
 
 tungsten, 546. 
 Estimation of tin by the humid method .... 546 
 
 Klaproth's process, 546. J. B. Hallet's process, 548. M. 
 
 Moissenet's process, 548. 
 Assay of tin in gun and bell metal .... 549 
 
 E. Burse's method, 550. 
 
 Estimation of tin by means of a standard solution . . 550 
 
 M. Gaultier de Claubry's process, 550. M. Lenssen's process, 
 551. M. Stromeyer's process, 551. Method for the analysis 
 of tin ore, 552. Another method, 553. Decomposition of 
 tin slags, 553. MM. Mohr and Terreil's method, 554. 
 Rammelsberg's method, 554. 
 
 CHAPTER XIII. 
 
 Assay of antimony . . . . . .556 
 
 Classification of antimonial substances .... 556 
 
 Class I. Native antimony and all antimonial substances con- 
 taining oxygen or chlorine, but little or no sulphur : 
 Native antimony, 556. Antimony oxide, 556. Antimo- 
 nious acid, 556. Antimonic acid, 556. 
 
 Class II. Antimony sulphide and all antimonial ores contain- 
 ing sulphur : Antimony sulphide, 556. Antimony oxy- 
 sulphide, 556. Haidingerite, 556. 
 Assay of ores of the first class ..... 556 
 
 Assay of ores of the second class . . . . .557 
 
 1. Estimation of the pure antimony sulphide (antimonium 
 crudum) . . . . . . . 557 
 
 2. Estimation of regulus of antimony . . . 558 
 
 Two methods, 558. 
 
 a. By roasting and fusing the oxidised matter with 
 
 black flux, 558. 
 
 b. By fusing the crude ore with iron, or iron scales, 
 
 with or without the addition of black flux, 558. 
 
 F. Becker's method, 563. Mr. Button's method, 563. 
 Detection of antimony in sublimates .... 563 
 
CONTENTS. [XXV] 
 
 PAGE 
 
 To distinguish arseniuretted hydrogen from antimoniuretted 
 
 hydrogen ....... 564 
 
 Separation of tin from antimony and arsenic . . . 564 
 
 0. Winckler's method, 564. 
 
 Assay of alloys of lead and antimony (type metal) . . 565 
 
 CHAPTER XIV. 
 
 Assay of zinc ....... 567 
 
 Classification of bodies containing zinc . . . .567 
 
 Class I. Zinc ores in which the metal exists "as oxide not 
 
 combined with silica, 567. 
 Class II. Zinc ores in which the metal exists as oxide, but 
 
 partly or wholly combined with silica, 567. 
 Class III. Zinc ores in which the metal is partly or wholly 
 
 combined with sulphur, 567. 
 Class IV. Alloys, 567. 
 Assay of ores of the first class . . . . .567 
 
 Assay of zinc by the humid process in ores of the first class . 571 
 Assay of ores of the second class . . . . .572 
 
 Wet assay of zinc in ores of the second class . . ,572 
 
 Assay of ores of the third class . . . . .573 
 
 Humid wet assay of zinc in ores of the third class . . 573 
 
 Assay of cupriferous blende ..... 574 
 
 Alloys (Class IY.) 574 
 
 Volumetric assay of zinc . . . . . .575 
 
 Galetti's process, 575. Schaffner's method, modified by C. 
 Kunzel, 577. 
 a. Solution of the ore, and preparation of the ammoniacal 
 
 solution, 577. 
 
 6. Preparation and standardising of the sodium sulphide 
 solution, 578. 
 
 c. Assay of the zinc in the solution of the ore, 579. 
 
 d. Further modification of the process, 580. 
 
 H. Schwarz's method, 581. Carl Mohr's method, 582. J. 
 
 Drewson's method, 583. 
 Separation of copper from zinc, 584. 
 
 CHAPTER XV. 
 
 Assay of mercury ....... 587 
 
 Assay of mercurial ores ...... 587 
 
 Berthier's method, 590. Eschka's method, 591. 
 
 Assay for the amount of cinnabar in an ore . . . 592 
 
 Electrolytic assay of mercury ..... 594 
 Volumetric estimation of mercury . . . .595 
 
 M. J. Personnel method, 595. M. Rivot's method, 597. 
 G. Attwood's method, 598. 
 
XXvil CONTENTS. 
 
 CHAPTER XVI. 
 
 PAGE 
 
 Assay of silver ....... 600 
 
 Classification of argentiferous substances . . . .600 
 
 Class I. Minerals containing silver, 600. 
 Class II. Metallic silver and alloys, 600. 
 General observations on the assay of ores and substances of 
 
 Class I. . . . . . .600 
 
 Fusion with oxidising reagents ..... 603 
 
 Litharge, 603. 
 Special directions for the crucible assay of ores and substances of 
 
 Class I. ....... 605 
 
 Preliminary assay for dividing all substances of this class 
 into three sections, 606. Assay of reducing power of 
 argol, 607. Assay of oxidising power of potassium 
 nitrate, 607. Assay of litharge for silver, 608. 
 Assay of ores of the first section ..... 608 
 
 Assay of ores of the second section . . . .609 
 
 Assay of ores of the third section ..... 609 
 
 Scorification . . . . . . .610 
 
 The roasting, 612. The fusion, 612. The scorification, 613. 
 Special instructions for the scorification assay of ores of Class I. . 615 
 
 Assay in scorifier, 615. 
 Assay of substances of Class I. admixed with native or metallic 
 
 silver ........ 616 
 
 Cupellation . . . . . . .617 
 
 Pliers and microscope, 623. 
 Amalgamation . . . . . . .626 
 
 Substances of the second class . . . . .627 
 
 Separation of silver from galena . . . . .627 
 
 General remarks on the assay of the alloys of silver and copper . 627 
 Cupellation, 627. D'Arcet's results, 628. Quantity of lead 
 required, 629. Loss of silver in the assay of coined 
 alloys, 630. 
 
 Special instructions for the assay of the alloys of silver and copper 631 
 Assay for approximative quantity of alloy, 631. Assay proper 
 of silver bullion, 631. Assay of alloys of copper and 
 silver, 632. 
 Alloys of platinum and silver . . . . .632 
 
 Alloy of platinum, silver, and copper .... 633 
 
 Assay of native silver, rough silver left on sieve during pulverisa- 
 tion of silver ores of first class, and native alloys of silver . 633 
 Dr. W. Dyce's process for separating gold and silver from the 
 
 baser metals ....... 633 
 
 Assay of silver bullion by the wet method . . .634 
 
CONTENTS. [xxvii] 
 
 PAGE 
 
 Gay-Lussac's method ...... 634 
 
 Measurement of the solution of common salt, 636. Measure 
 of the normal solution of salt by weight, 636. Preparation 
 of the decime solution of common salt, 638. Preparation 
 of the decime solution of silver, 640. Weighing the normal 
 solution of common salt, 641. Preparation of the normal 
 solution of common salt when measured by weight, 641. 
 Preservation of the normal solution of common salt, 646. 
 Application of the process described in the determination 
 of the standard of a silver alloy, 647. Correction of the 
 standard of the normal solution of salt when the tempera- 
 ture varies, 649. Table of corrections for variations in 
 temperature of the normal salt solution, 651. 
 
 Table for the assay, by the wet method, of an alloy containing any 
 proportions whatever of silver, by the employment of a con- 
 stant measure of the normal solution of common salt . 651 
 Tables for determining the standard of any silver alloy by employ- 
 ing an amount of alloy always approximatively containing 
 the same amount of silver . . . . 654, 673 
 
 Assay of pure, or nearly pure silver, the temperature of the 
 normal solution of salt being that at which it was standard- 
 ised ........ 674 
 
 First example, 674. Second example, 675. Third example, 
 
 676. 
 Graduation of the normal solution of salt, the temperature being 
 
 different to that at which it was wished to be graduated . 677 
 
 First example, 677. Second example, 678. 
 Approximative determination of the standard of an unknown 
 
 alloy ........ 678 
 
 Modes of abridging manipulation . . . . .679 
 
 Bottles, 679. Stand, 680. Water-bath, 680. Flue, 681. 
 
 Agitator, 681. Table, 683. Cleaning the bottles, 684. 
 Reduction of silver chloride obtained in the assay of alloys by the 
 
 wet method . . . . . . . 684 
 
 Preparation of pure silver . . . . . .685 
 
 Modifications required in the assay of silver alloys containing 
 
 mercury ....... 68fr 
 
 Method of taking the assay from the ingot, 687. Method of 
 assaying silver bars adopted in the assay offices of H.M. 
 Indian mints, 687. The chloride process, 688. Apparatus 
 and appliances required, 697. 
 
 Effect of bismuth on the ductility of silver . . . 699' 
 
 Titration o^ silver in presence of other metals . . .711 
 
 Copper, 713. Mercury, 714. Palladium, 714. 
 
 Determination of silver in galena by Volhard's process . .714 
 
 David Forbes on blowpipe assay of silver . . . .715- 
 
 Apparatus, 716. Concentration of the silver lead, 717. Cu- 
 
[xxviii] CONTEXTS. 
 
 PAGE 
 
 .pellation, 720. Estimation of the weight of the silver 
 globule obtained in cupellation, 723. The scale for this 
 purpose, 724. Cupellation loss, 727. Table modified by 
 Plattner, 728. 
 Classification of argentiferous substances . . . .729 
 
 A. Metallic alloys capable of direct cupellation, 730. 
 
 a. Consisting chiefly of lead or bismuth, 730. 
 
 6. Consisting chiefly of silver, and alloys of silver with 
 
 gold and copper, 731. 
 c. Consisting chiefly of copper, 732. 
 
 B. Metallic alloys incapable of direct cupellation, 733. 
 
 a. Containing much copper or nickel, with more or less 
 
 sulphur, arsenic, etc., 733. 
 
 b. Containing tin, 734. 
 
 c. Containing antimony, tellurium, or zinc, 735. 
 
 d. Containing mercury ; amalgams, 737. 
 
 e. Containing much iron, 737. 
 Alloys of silver and copper, 739. 
 
 CHAPTER XVII. 
 
 Assay of gold ....... 740 
 
 Classification of substances containing gold . . .740 
 
 Class I. Ores containing gold, 740. Graphic tellurium, 740. 
 
 Folliated tellurium, 740. 
 Class II. Alloys of gold, 740. Native gold and aurides of 
 
 silver, 740. Artificial alloys of gold, 740. 
 Assay of gold ores . ...... 741 
 
 Preparation of the sample, 742. Collection of the gold and 
 silver, 742. Crucible assay, 742. The charge, 743. Size 
 of lead button, 743. Preliminary assay of ore, 744. 
 Roasting the ore, 745. Fusion, 747. Scorification assay, 
 747. The lead button, 749. Cupellation, 749. 
 
 Estimating the weight of minute spheres of gold . . . 753 
 
 Oeneral observations on the assay of gold ores . .. .756 
 
 Gold and copper, proportion of lead, 756. Examination on 
 the touchstone, 757. Table for the proportion of lead to 
 be employed in the cupellation of gold and copper, 759. 
 Gold, silver, platinum, and copper, 759. Gold alloyed with 
 silver, 761. Inquartation, 761. Surcharge, 763. Mr. 
 Seine's method, 763. Makin's method, 765. Aqua regia, 
 766. Yon Jiiptner's method, 767. 
 Standard of the alloys of gold . . . . .768 
 
 Assay of gold coin and bullion . . . . .769 
 
 Preliminary assay, 769. Assay proper, 769. Parting assays, 
 771 
 
CONTENTS. 
 
 PAGE 
 
 Assay of pyrites for gold . . . . . .773 
 
 Treatment of gold and silver bearing copper ores . . .775 
 
 Detection of minute traces of gold in minerals . . . 777 
 
 CHAPTER XVIII. 
 
 Assay of platinum . . . . . . .781 
 
 Platinum in its native state . . . . .781 
 
 Analysis of platinum ores . . . . . .782 
 
 Bunsen's method, 782. Platinum and palladium, 782. Ru- 
 thenium, rhodium, and iridium, 785. C. Lea's process, 
 790. Deville and Debray's process, 795. Sand, 795. Osm- 
 iridium, 796. Platinum and iridium, 797. Palladium, iron, 
 and copper, 798. Gold and platinum, 799. Rhodium, 799. 
 Analysis of platinum ores from various sources, 799. 
 Guyard's process for extracting metals from platiniferous 
 residues, 799. Analysis of osm-iridium, 803. Wolcott 
 Gibbs's process, 804. Nelson Perry's process, 807. 
 
 CHAPTER XIX. 
 
 Assay of bismuth ....... 809 
 
 Varieties of bismuth ores ...... 809 
 
 Native bismuth . . . . . .809 
 
 Assaying bismuth ores . . . . , .810 
 
 Assaying bismuth in ores containing a large amount of copper . 810 
 Refining crude bismuth . . . . . .813 
 
 Purification of the reduced bismuth . . . .814 
 
 Purification of bismuth from arsenic, 814. Purification of 
 bismuth from antimony, 815. Purification of bismuth 
 from copper, 815. Purification of bismuth from sulphur, 
 816 
 Volumetric assay of bismuth . . . . .817 
 
 R. W. Pearson's process, 817. M. P. Muir's process, 818. 
 
 CHAPTER XX. 
 
 Assay of chromium ...... 820 
 
 Principal ore of chromium . . . . . 820 
 
 Assay of chrome ore ...... 820 
 
 Genth's process, 820. O'Neill's process, 822. W.Gibbs's pro- 
 cess, 823. Clark's process, 824. H. N. Morse and W. C. 
 Day's process, 825 
 
 Estimation of chromium by means of standard solutions . . 827 
 
 Estimation of chromium in iron and steel, 827. J. O. 
 Arnold's process, 827. W. J. Sell's process, 830. 
 
[XXX] CONTENTS. 
 
 CHAPTER XXI. 
 
 PAGE 
 
 Assay of arsenic . . . . . . .831 
 
 Minerals from which arsenic is produced . . . .831 
 
 Assay for arsenic . . . . . . 831 
 
 Approximative method, 832. Mr. Parnell's method, 832. 
 
 CHAPTER XXII. 
 
 Assay of manganese . . . . . .833 
 
 Commercially valuable minerals containing manganese . .833 
 
 Valuation of manganese ores . . . . .833 
 
 Sherer and Rumpf 's method, 834. Mohr's method, 837. Otto's 
 
 method, 837. Bunsen's method, 837. J. Pattinson's 
 
 process, 837. 
 
 CHAPTER XXIII. 
 
 Assay of nickel and cobalt ores . . . . .839 
 
 Ores of nickel . . . . . . .839 
 
 Ores of cobalt . . . . . . .839 
 
 Hadow's process for separating nickel and cobalt . . .839 
 
 Decomposition of cobalt speiss . . . . .843 
 
 Assay of nickel ores ...... 845 
 
 Assay of commercial metallic nickel .... 845 
 
 Assay of cobalt ores ...... 846 
 
 Liebig's method of separating nickel and cobalt . . . 846 
 
 Wolcott Gibbs's improvement . . . . .847 
 
 Terreil's method . . . . . . .847 
 
 Fleitmann's quantitative assay of small proportions of cobalt in 
 
 nickel. ....... 849 
 
 Plattner's method for detecting nickel before the blowpipe . 849 
 
 Assay of complex nickel and cobalt ores .... 850 
 
 A. Olassen's method, 850. Ores containing sulphur, arsenic, 
 
 nickel, cobalt, and iron, 851. Alloys of zinc, copper, and 
 
 nickel, 853. 
 
 Ohl's method for the assay of nickel speiss . . .854 
 
 Lead and copper speiss, 855. Speiss containing little or no 
 
 lead or antimony, 855. Speiss containing much lead or 
 
 antimony, 856. 
 
 CHAPTER XXIV. 
 
 Assay of sulphur ....... 859 
 
 Commercially valuable sulphur minerals . . . .859 
 
 Assay of sulphur in iron and copper pyrites . . .859 
 
CONTENTS. [xxxi] 
 
 PAGE 
 
 Assay of sulphur in the dry way . . . . .859 
 
 Assay of sulphur in the wet way . . . . .861 
 
 C. R. A. Wright's process, 861. Pearson's method, 862. 
 
 Hougeau's process, 864. Deutecom's method, 864. 
 
 Holland's method, 865. Breckmann's process, 866. 
 
 CHAPTER XXV. 
 
 Discrimination of gems and precious stones . . . 867 
 
 Explanation and introduction, 867. Principal sources of 
 
 recognition, 867. 
 Colourless stones . . . . . . .867 
 
 Diamond, 867. The matrix of the diamond, 868. Quartz, 
 874. White zircons, 875. White sapphire, 876. White 
 topaz, 876. Comparative table of the weights of colour- 
 less stones, 878. Use of the tatye, 878. 
 Yellow stones . . . . . . .879 
 
 Yellow zircon, 879. Yellow sapphire, 879. Cymophane 
 (chrysoberyl), 879. Yellow topaz, 880. Yellow tourmaline, 
 880. Yellow emerald, 881. Comparative table of weights 
 of yellow stones, 883. Yellow quartz, 884. 
 Brown and flame-coloured stones ..... 884 
 
 Zircon (hyacinth), 884. Yermeil garnet, noble garnet, al- 
 mandine, 884. Comparative table of weights of brown or 
 flame-coloured stones, 885. Essonite, cinnamon stone, 886. 
 Tourmaline, 886. 
 Red and rose-coloured stones ..... 886 
 
 Red sapphire, oriental ruby, 886. Deep red garnet, noble 
 garnet, 886. Spinel ruby, 886. Reddish topaz, 887. Red 
 tourmaline, 887. Comparative table of weights of red and 
 rose-coloured stones, 887. 
 Blue stones ....... 888 
 
 Blue sapphire, 888. Disthene, cyanite, 888. Blue topaz, 888. 
 Blue tourmaline, 889. Blue beryl, 889. Dichroite, water 
 sapphire, 889. Turquoise, 889. Comparative table of the 
 weights of blue stones, 890. 
 
 Violet stones .890 
 
 Violet sapphire, 890. Violet tourmaline, 890. Violet 
 quartz, amethyst, 890. Comparative table of weights of 
 violet stones, 891. 
 
 Green stones . . . . .891 
 
 Green sapphire, 891. Peridot, crysolite, 891. Green tourma- 
 line, 892. Emerald, 892. Aqua-marine, 892. Chryso- 
 prase, 892. Comparative table of weights of green stones, 
 896. 
 
[xxxii] 
 
 CONTENTS. 
 
 Stones possessing a play of colours (chatoyant) . 
 
 Sapphire, 894. Garnet, 894. Cymophane, 894. Antique 
 
 emerald, 894. Quartz, 894. Felspar, nacreous felspar, 
 
 fish-eye, &c., 895. Comparative table of weights of stones 
 
 possessing a play of colours (chatoyant), 895. 
 
 Glass and artificial gems ...... 
 
 Inferior brilliancy, 896. Inferior hardness, 896. Fusibility, 89 6. 
 
 PAGE 
 
 893 
 
 896 
 
 APPENDIX. 
 
 TABLE I. Showing the quantity of fine gold in 1 oz. of any 
 alloy to ^ of a carat grain of the mint value 
 of 1 oz. of each alloy . . ii 
 
 TABLES A, B, and C. To convert mint value into bank 
 value when the standard is expressed in 
 carats, grains, and eighths ... xx 
 
 TABLE II. Table of relative proportions of fine gold and 
 alloy, with the respective mint values of 
 1 oz. of each alloy when the standard is 
 expressed in thousandths . . . xxi-xxxii 
 
 TABLE. To convert mint value into bank value when the 
 
 standard is expressed in thousandths . . xxxiii 
 
 TABLE III. Assay table, showing the amount of gold and 
 silver, in ounces, pennyweights, and grains, 
 contained in a ton of ore, &c., from the 
 weight of metal obtained in an assay of 200 
 grains of mineral .... xxxiv-xlvii 
 
 INDEX 
 
 xlix 
 
. 
 A MANUAL 
 
 OF 
 
 PEACTICAL ASSAYING, 
 
 CHAPTEE I. 
 
 CHEMICAL NOMENCLATURE LAWS OF COMBINATION, ETC. 
 
 IN a treatise intended to be used principally by the prac- 
 tical assayer, it is neither necessary nor possible to give 
 more than a brief outline of the elements of chemical 
 nomenclature and of chemical combination. A knowledge 
 of practical chemistry is undoubtedly of great value to 
 the assayer ; indeed, no one can attain to any degree of 
 eminence in this branch of industry unless he has had 
 some amount of practice in the laboratory ; but it will be 
 beyond the scope of this volume to teach the elements of 
 chemistry. Such instruction in chemistry must be sought 
 for in books which are specially devoted to the science. 
 The student should, above all, endeavour to acquire a 
 practical knowledge of experimental chemistry by going 
 through a course of instruction in a laboratory. 
 
 CHEMICAL NOMENCLATURE. Every material substance with 
 which we are acquainted consists of one or more bodies, 
 termed elements, from the fact that with our present means 
 of research we are unable to reduce them to a more simple 
 form. Thus, if a piece of common iron pyrites be exposed 
 
 B 
 
L> CHEMICAL NOMENCLATURE 
 
 to certain chemical operations, it will be found to consist 
 of two substances, each physically and chemically distinct 
 from the other and from the original substance. One 
 body is sulphur, an opaque yellow substance, fusing at a 
 very low temperature, igniting readily, and burning with 
 a peculiar suffocating odour. The other constituent is iron, 
 a well-known metallic substance, requiring an intense heat 
 for fusion, and not burning at a red heat. If we perform 
 any experiment which, in the present state of knowledge, 
 ingenuity could suggest, we are totally unable to cause 
 either the sulphur or the iron to assume a more simple or 
 elementary state of existence. We can with ease cause 
 either of them to enter into new combinations with other 
 bodies, and these compounds we can decompose as in the 
 case of the pyrites and obtain both sulphur and iron again 
 in their separate forms with all their characteristic proper- 
 ties ; but nothing more than this can be effected : hence 
 we are led to the belief that both sulphur and iron are 
 elements, or bodies containing only one kind of matter. 
 
 The following table gives the elements discovered up 
 to the present time. Those substances whose names are 
 printed in italics have hitherto been found of no practical 
 use ; and those marked with an asterisk (*) are often 
 found native, or unassociated with mineralising elements. 
 
 Names of the 
 Elements 
 
 Aluminium 
 
 Antimony 
 
 Arsenic 
 
 Barium 
 
 Beryllium 
 
 * Bismuth 
 Boron 
 Bromine 
 Cadmium 
 Ctesium 
 Calcium 
 
 * Carbon 
 Cerium 
 Chlorine 
 
 Elements. 
 
 - - 
 
 
 i 
 
 Symbols 
 
 Atomic 
 Weights 
 
 Names of the 
 Elements 
 
 Al 
 
 27 
 
 Chromium 
 
 Sb 
 
 119-6 
 
 Cobalt 
 
 As 
 Ba 
 
 74-9 
 136-9 
 
 *Copper 
 Didymium 
 
 Be 
 
 9 
 
 Erbium . 
 
 Bi 
 
 207-5 
 
 Fluorine . 
 
 B 
 
 11 
 
 Gallium . 
 
 Br 
 
 80 
 
 Germanium 
 
 Cd 
 
 112 
 
 *Gold . 
 
 Cs 
 
 132-7 
 
 Hydrogen . 
 
 Ca 
 
 40 
 
 Indium 
 
 C 
 
 12 
 
 Iodine 
 
 Ce 
 
 141-2 
 
 *Iridium 
 
 Cl 
 
 35-5 
 
 Iron . 
 
 Symbols 
 
 Cr 
 
 Co 
 
 Cu 
 
 Di 
 
 E 
 
 F 
 
 Ga 
 
 Ge 
 
 Au 
 
 H 
 
 In 
 
 I 
 
 Ir 
 
 Fe 
 
 A.tomic 
 Weights 
 
 52-5 
 58-6 
 63-5 
 146 
 166 
 19 
 
 69-9 
 
 72-3 
 
 196-2 
 
 1 
 
 113-4 
 
 126-5 
 
 192-5 
 
 56 
 
CHEMICAL NOMENCLATURE. 
 
 Elements cont. 
 
 "ssas" ! s "" b <"> 
 
 Atomic 
 Weights 
 
 Names of the 
 
 Elements 
 
 Symbols 
 
 Atomic 
 Weights 
 
 Lanthanum La 
 
 138-5 
 
 Selenium . 
 
 Se 
 
 78-9 
 
 Lead. . Pb 
 
 207 
 
 Silicon 
 
 Si 
 
 28 
 
 Lithium . Li 
 
 7 
 
 *Silver 
 
 Ag 
 
 108 
 
 Magnesium j Mg 
 
 24 
 
 Sodium 
 
 Na 
 
 23 
 
 Manganese : Mn 
 
 54-8 
 
 Strontium 
 
 Sr 
 
 87-3 
 
 Mercury . Hg 
 
 199-8 
 
 * Sulphur 
 
 S 
 
 32 
 
 Molybdenum Mo 
 
 95-9 
 
 Tantalum 
 
 Ta 
 
 182 
 
 Nickel . ! Ni 
 
 58-6 
 
 Tellurium 
 
 Te 
 
 127-7 
 
 Niobium (Co 
 
 
 Terbium . 
 
 Tb 
 
 ? 
 
 lumbium, Cb) Nb 
 
 93-7 
 
 Thallium . 
 
 Tl 
 
 203-7 
 
 Nitrogen . N 
 
 14 
 
 Thorium . 
 
 Th 
 
 232 
 
 Osmium . Os 
 
 195 
 
 Thulium . 
 
 Tm 
 
 ? 
 
 Oxygen . O 
 
 16 
 
 Tin . 
 
 Sn 
 
 118 
 
 Palladium . j Pd 
 
 106-2 
 
 Titanium . 
 
 Ti 
 
 50-2 
 
 Phosphorus \ P 
 
 31 
 
 Uranium . 
 
 U 
 
 239-8 
 
 Platinum . | Pt 
 
 194-3 
 
 Vanadium 
 
 V 
 
 51-1 
 
 Potassium K 
 
 39 
 
 Wolfram(Tung 
 
 
 
 Rhodium . i Rh 
 
 104-1 
 
 sten) 
 
 W 
 
 183-6 
 
 Rubidium Rb 
 
 85-2 
 
 Ytterbium 
 
 Yb 
 
 172-6 
 
 Ruthenium Ru 
 
 103-5 
 
 Yttrium . 
 
 Y 
 
 89-6 
 
 Samarium j Sa 
 
 150 
 
 Zinc . 
 
 Zn 
 
 65 
 
 Scandium '< Sc 
 
 44 
 
 Zirconium 
 
 Zr 
 
 90-4 
 
 The first column contains the name of the element ; 
 the second, the symbol, in which all chemical changes and 
 decompositions are most readily understood ; and the third, 
 the atomic weight. These atomic weights are not given 
 beyond the first place of decimals, to avoid tedious calcu- 
 lation ; for all practical purposes they may be considered 
 accurate. Of the compounds of these elements, only those 
 will be discussed which are likely to fall under the notice 
 of the assayer. 
 
 The principal compound bodies with which the assayer 
 will have to deal are acids, oxides, salts, and binary, sub- 
 stances containing no oxygen. 
 
 When a body combines in more than one proportion 
 with oxygen, that compound containing the least oxygen 
 takes the termination ous^ that containing the most ic\ 
 thus,: sulphurous acid, sulphuric acid ; arsenious acid, 
 arsenic acid ; ferrous oxide, ferric oxide ; mercurous 
 
 u -2 
 
CHEMICAL NOMENCLATURE. 
 
 oxide, mercuric oxide ; in only one proportion (or 
 when they form only one basic oxide) they are distin- 
 guished by the termination ic, as potassic oxide, aluminic 
 oxide. 
 
 OXIDES are binary oxygen compounds; they may be 
 divided into three series. The first series comprises those 
 oxides which do not possess the property of combining with 
 acids to form salts they are termed indifferent oxides ; 
 the second contains those capable of uniting with acids to 
 form salts, and called salifiable oxides or bases ; the third 
 comprises those oxides which have acid characters, and 
 form salts by uniting with bases. 
 
 When an elementary body combining with oxygen 
 forms but one oxide, it is simply called the oxide of that 
 element. Thus we say zinc oxide, potassium oxide, alumi- 
 nium oxide. 
 
 If the body is capable of combining with oxygen in 
 many proportions, the words proto-, sesqui-, bin-, or per-, 
 &c., precede the term oxide, to express the progressive 
 amounts of oxygen. Most metals form one salifiable 
 oxide, and many of them have two ; these are now gener- 
 ally distinguished by the terminations ous and ic, in the 
 same manner as are the acids. Thus we have protoxide of 
 lead, iron, copper, tin, &c. ; sesquioxide of aluminium, iron, 
 or chromium, &c. ; binoxide or peroxide of manganese, 
 copper, mercury, &c. ; and when we speak of them as 
 salifiable bases, ferrous and ferric oxides ; mercurous and 
 mercuric oxide. Some metals unite with oxygen in still 
 higher proportions ; these compounds are almost always 
 acids, such as chromic acid, stannic acid, antimonic 
 acid, &c. 
 
 SALTS are formed when an acid unites with a base, and 
 usually the properties of the acid and the base are recipro- 
 cally neutralised ; thus an acid which before combination 
 possesses the power of reddening blue litmus, loses it on 
 combining with the base, and, in like manner, a base which 
 would at first change reddened litmus paper to blue loses 
 this property as the acid saturates it. In this case the 
 acid and base have combined to form a salt. 
 
LAWS OP COMBINATION. 5 
 
 The names of salts are governed first by the nature of 
 the acid ; secondly, by the salifiable nature of the base ; 
 and, thirdly, by the proportions in which the acid and base 
 are combined. Acids terminating in ic form salts ending 
 in ate. Acids terminating in ous form salts terminating in 
 ite ; and the new names having these terminations are 
 added to the name of the oxide. Thus sulphuric acid and 
 iron protoxide form sulphate of iron protoxide, ferrous 
 sulphate, or, more commonly, iron protosulphate ; arseni- 
 ous acid and iron protoxide form arsenite of iron prot- 
 oxide, ferrous arsenite, or iron protarsenite ; nitric acid 
 and iron sesquioxide form nitrate of iron sesquioxide or 
 ferric nitrate. 
 
 When the salt exists in the neutral state its name is 
 .formed as above, but if the proportion of acid is greater 
 than in neutral salts, it is termed an acid salt: thus 
 potassium bisulphate is sometimes called acid sulphate of 
 potassium. If, on the other -hand, the base is in excess, 
 the name is preceded by the words sub or basic ; thus, 
 lead subacetate or basic- acetate of lead. 
 
 Binary compounds containing no oxygen exist very 
 largely in nature, and it is from them that the greater 
 part of our copper, lead, silver, &c., is obtained. 
 
 When a non-metal combines with a metal to. form 
 a compound which is neither acid nor basic, its name 
 is derived from the non-metal by the addition of the 
 termination uret or ide. The latter term is, however, 
 gradually displacing the former. Thus the compounds 
 of sulphur with iron and chlorine with silver are res- 
 pectively called iron sulphuret or sulphide, and silver 
 chloride. 
 
 If a non-metal combines with a metal in more than 
 one proportion, the same rule is followed as with the 
 oxygen compounds : thus we have ironjpn>fo-sulphide, iron 
 sesqui-sulphide, andiron Si-sulphide (ordinary iron pyrites 
 or mundic). 
 
 Laws of Combination. On examining the compounds 
 which the same substances afford by their union in dif- 
 ferent proportions, it has been noticed that the propor- 
 
(5 LAWS OF COMBINATION. 
 
 tions of the elements existing in each compound are 
 definite ; a certain weight of one substance will only com- 
 bine with a certain weight of another substance, and the 
 lowest combining weight of any of the elementary bodies 
 is termed its atomic weight, and is represented by the 
 numbers in the third column of the table of elementary 
 substances. 
 
 As before stated, all substances combine in fixed or 
 definite proportions ; thus, if 223 parts of oxide of lead 
 are analysed, they will be found to consist of 207 parts of 
 lead and 16 of oxygen. Again, the analysis of 18 parts of 
 water or oxide of hydrogen would give 2 parts of hydro- 
 gen and 1 6 of oxygen ; now, taking hydrogen as unity, we 
 have 207 as the equivalent of lead, and 16 as that of 
 oxygen. If we follow oxygen further in its combinations, 
 it will be seen that' 
 
 16 parts of oxygen combine with 1 part of hydrogen. 
 
 207 lead. 
 
 40 calcium. 
 
 , 118 tin. 
 
 63-5 copper. 
 
 The above numbers, therefore, represent the equivalents 
 of the respective elements. 
 
 Again, the equivalent of sulphur is 32, and this 
 represents the weight of sulphur which will combine with 
 the above weights of hydrogen, lead, calcium, tin, or 
 copper to form sulphides of the respective bases. 35*5 
 parts of chlorine, or 78*9 parts of selenium, also combine 
 with the same weights, viz. hydrogen 1, lead 207, &c., 
 to form chlorides and selenides. 
 
 Compounds like these are of the simplest class, and 
 consist of single equivalents only ; there are, however, 
 many compounds containing more than two equivalents, 
 in which case the following laws are followed. 
 
 In one class of compounds the quantity of one of the 
 constituent elements remains constant, while each new 
 compound is formed by the successive addition of another 
 
CHEMICAL SYMBOLS. 7 
 
 atom of the other constituent element ; and it must also 
 be borne in mind that no element will combine with 
 another in less than its atomic weight. Another series 
 will commence with two atoms of an element united with 
 an uneven number of atoms of another element ; thus 
 we can have binary compounds in the proportion of 2 to 
 3, 2 to 5, or 2 to 7. 
 
 The atomic weight of a compound body is the sum 
 of the atomic weights of the elements forming it : thus 
 sulphuric anhydride is composed of one atom or 32 parts 
 of sulphur, and 3 atoms, or 48 parts, of oxygen ; its atomic 
 weight is therefore 80. The atomic weight of any com- 
 pound body may be ascertained by adding together the 
 atomic weights of its constituent elements. 
 
 Owing to the invariable law of the constancy of chemi- 
 cal compounds, we are enabled to calculate the reaction 
 which occurs between two or more bodies when decompo- 
 sition takes place: thus 174 parts of potassium sulphate 
 contain 80 parts of sulphuric anhydride and 94 parts 
 of potassium oxide ; and if it were desired to obtain lead 
 sulphate by the decomposition of lead nitrate by adding 
 to it the above quantity of potassium sulphate, the exact 
 amount of lead nitrate required would be readily found by 
 adding together the equivalent of the elements forming 
 nitric acid and lead oxide. 
 
 CHEMICAL SYMBOLS : THEIR EMPLOYMENT AND USES The 
 symbol of an element standing alone signifies one atom 
 of that element. Thus : S implies not only the element 
 sulphur, but 32 parts of sulphur ; a small figure on the 
 right-hand side of the symbol indicates the number of 
 atoms to be represented ; thus, S 2 is equal to two atoms, 
 or 64 parts of sulphur. 
 
 Two symbols placed thus, FeS, indicate a compound 
 of equal equivalents of iron and sulphur. Separation 
 of elements by the sign + or a comma is employed to 
 show the union of two compound bodies ; thus the com- 
 pound of silver sulphide and lead sulphide may be thus 
 written : AgS + PbS, or AgS,PbS. A large figure on the 
 same line as the symbol, and on its left side, multiplies the 
 
CHEMICAL SYMBOLS. 
 
 whole of the symbols to the first comma or + sign : thus, 
 2AgS,PbS, or 2AgS + PbS, represents a compound of 
 two atoms of silver sulphide with one of lead sulphide. 
 If, however, it be thus written, 2(AgS, PbS), it means two 
 atoms of the whole of the elements which are inclosed in 
 the brackets. 
 
PREPARATION OP THE SAMPLE. 
 
 CHAPTER II. 
 PREPARATION OF THE SAMPLE WEIGHING. 
 
 THE selection and preparation of the sample is the first 
 and most important 'operation in assaying. It is of little 
 use for the operator to ascertain with accuracy the per- 
 centage of every individual constituent in the mineral 
 operated on, if the sample does not truly represent the 
 average of the ore. It should be borne in mind that 
 samples of mineral are generally selected for their richness, 
 and represent the most favourable portions of the ore ; 
 and no pains should be spared to secure a sample for 
 analysis which will truly show the bulk of mineral whose 
 value is required to be known. 
 
 The assayer must always bear in mind the object which 
 his experiments have in view. If they are to ascertain 
 the actual percentage of one or more constituents existing 
 in a certain stone, his labours are comparatively easy, all 
 that is required being to reduce the whole of the specimen 
 to the finest possible state of division, and, having well 
 mixed the powder, to analyse a portion of it. 
 
 But if it is desired to find out the composition of a 
 special mineral or crystal, the greatest possible care must 
 be taken to remove the whole of the gangue or other im- 
 purities, and to obtain for analysis those portions only 
 which represent with greatest accuracy the pure mineral. 
 To effect this the surrounding rock is first removed as 
 carefully as possible, and then the specimen is crushed 
 into coarse pieces on a sheet of clean paper. By means 
 of a pocket magnifier and a pair of pincers, clean, typical 
 pieces of the mineral are then to be selected for analysis. 
 
10 PKEPARATIOX OF THE SAMPLE. 
 
 If, however, as will most frequently be the case, the 
 object of the assay be to ascertain the average value of a 
 mineral lode or heap of ore, then the assayer must proceed 
 differently. The portion experimented upon must truly 
 represent, in the respective amounts of its valuable ma- 
 terial, gangue, quartz, and earthy matters, the whole bulk 
 of that of which it professes to be a sample ; and this 
 having been secured, the whole must be carefully pow- 
 dered and passed through fine sieves, taking care that 
 every portion of the mineral goes through. If this be not 
 attended to, it will frequently happen that the few grains 
 left out are sufficient to vitiate the whole assay ; this is 
 especially apt to be the case when examining ores the 
 valuable ingredients of which are of a ductile or malle- 
 able nature, such as auriferous quartz. In this case it fre- 
 quently happens that the great, bulk of gold exists in the 
 form of one or two small pieces, and these being flattened 
 and beaten out in the operation of powdering will almost 
 certainly be left upon the sieve. In cases like this it is 
 better to collect and assay such pieces separately, and esti- 
 mate their proportion to the whole weight of the sample, 
 than to attempt to powder and distribute them uniformly! 
 
 SAMPLING. The important operation of sampling neces- 
 sarily precedes any process of assay. Dr. Peters, in his 
 ' Modern American Methods of Copper Smelting,' gives the 
 following useful description of the operation of sampling 
 as practised in the great mining centres of the New World. 
 By sampling we seek to obtain within the compass of a few 
 ounces a correct representative of the entire quantity of 
 ore, which may vary in amount from a few pounds to 
 several thousand tons. As a rule it will lessen the chance 
 of serious error, in very large transactions, to divide the lot 
 into parcels of not over fifty tons each, and sample each 
 of these lots by itself. The utmost care and vigilance 
 in sampling and assaying should be required at every 
 smelting works, both in the interest of the works and in 
 that of the ore-seller. 
 
 Until quite recently, it has been customary to sample 
 lots of ore by quartering them down, rejecting a certain 
 
PREPARATION OF THE SAMPLE. 11 
 
 proportional part at each successive operation, and re- 
 ducing the size of the ore fragments as the quantity to 
 operate on diminishes. This is a laborious and expensive 
 method, and in the case of finely pulverised ores may 
 well be replaced by the use of the ' split shovel,' or one of 
 the many automatic sampling machines that have been 
 invented. 
 
 But since the establishment of public sampling works 
 at most of our great mining centres, where the correctness 
 of the sample is guaranteed by the works, which dis- 
 tribute packages of each lot of ore to the agents of the 
 various rival smelting companies for them to assay and 
 bid upon, the vast quantities of ores handled, and the im- 
 portance in many instances of retaining the lump form of 
 the ore as essential to the subsequent metallurgical opera- 
 tions, have imperatively demanded some method of auto- 
 matic sampling that shall be rapid, accurate, and equally 
 applicable to ores in both the pulverised and lump form. 
 
 The means hitherto employed all depend upon the 
 same general principle of cutting or dividing a falling 
 stream of ore by means of flanges, fingers, ore travelling 
 buckets, in such a manner as to obtain a certain desired 
 proportion of it for a sample. 
 
 While many of these devices work admirably upon 
 pulverised ore, free from dampness or foreign obstructing 
 substances, they are apt to give entirely unsatisfactory 
 results upon a mixture of fine and coarse ores, while the 
 presence of strings, chips, rags, &c., usually clogs them 
 and deranges their working. 
 
 Mr. D. W. Brunton, of Denver, Colorado, has invented 
 an automatic sampling machine that is apparently free 
 from all the defects enumerated, and which has been 
 shown by practical trial to be equally applicable to coarse, 
 fine, or mixed ores, while it cannot be clogged by foreign 
 bodies of any reasonable size.* 
 
 Brunton overcomes these difficulties by deflecting the 
 entire ore stream to the right or left, while falling through 
 
 * See Transactions of the American Institute of Mining Engineers, vol. 
 xiii. p. 639, for drawings and full description of this sampler. 
 
12 PREPARATION OF THE SAMPLE. 
 
 a vertical or inclined spout. By a simple arrangement of 
 movable pegs, in connection with the driving gear, the 
 proportion of the ore stream thus deflected into the 
 sample bin may vary from 10 to 50 per cent. ; the latter 
 amount only being required in coarse ores of enormous 
 and very variable richness, while for ordinary lump ores 
 from 10 to 20 per cent, is the maximum required. 
 
 Instead of passing the sample-stream of ore into a bin, 
 this system may be still further perfected by leading it 
 directly to a pair of moderately fine rolls, the product of 
 which is elevated to a second similar sampling-machine 
 from which the final sample drops into a locked bin. 
 
 Six months' constant experience with this sampler has 
 shown that 10 per cent, of 20 per cent., or 2 per cent, of 
 the original ore parcel is usually quite sufficient ; though 
 in exceptional cases, 15 per cent, of 30 per cent., or \\ 
 per cent, of the ore may be required. 
 
 The two machines are driven at different speeds, to 
 prevent any possible error that might arise from isochronal 
 motion, and by careful tests of this machine in resampling 
 lots of ore the limit of error has been found less than 
 one-fourth of 1 per cent. ; while even the best hand- 
 sampling may vary 2 per cent. 
 
 The fact that the division is one of time and not of ore 
 is one of the most important features of this valuable in- 
 vention, as it consequently is forced to deflect the exact 
 proportion of the ore-stream, for which it is set, whether 
 coarse or fine, wet or dry, Hght or heavy. 
 
 MOISTURE. The determination of the moisture present 
 in any given parcel of ore is also a matter of much im- 
 portance ; and probably more inaccuracies attend this 
 apparently simple process than any other of the pre- 
 liminary operations. 
 
 This determination must, of course, take place as nearly 
 as possible at the same time that the entire ore parcel is 
 weighed, as otherwise the sample may lose or gain 
 moisture. 
 
 In lump ores, it is difficult to obtain a correct sample 
 even for moisture, without some preliminary crushing, 
 
PREPARATION OF THE SAMPLE. 13 
 
 and to save labour, it is best to use a portion of the regu- 
 lar assay sample for this purpose ; the accurate weighing 
 of the entire ore parcel being postponed until just before 
 or after the sampling, and the portion reserved for the 
 moisture determination being placed in an open tin vessel, 
 contained in a covered metal case, having an inch or two 
 of water on its bottom, in which sample tins stand. 
 
 From one-fourth to one-half pound of the sample is 
 usually weighed out for this determination, and dried 
 under frequent stirring, and at a temperature not exceeding 
 212. While it is always important to keep within the 
 limit of temperature just mentioned, it is especially the 
 case with certain substances which oxidise easily. Among 
 these are finely divided sulphides, and, above all, the 
 pulverulent copper cements obtained from precipitating 
 copper with metallic iron from a sulphate solution. 
 
 Such a sample, containing actually 5J per cent, of 
 moisture, showed an increase of weight of some 2 per 
 cent., on being exposed for thirty minutes to a tempera- 
 ture of about 235 Fahr. 
 
 Certain samples of ore especially from the roasting 
 furnace are quite hygroscopic, and attract water rapidly 
 after drying. 
 
 In such cases the precautions used in analytical work 
 must be employed, and the covered sample weighed 
 rapidly in an atmosphere kept dry by the use of strong- 
 sulphuric acid. 
 
 The sampling of the malleable products of smelting, 
 such as blister copper, metallic bottoms, ingots, &c., can 
 only be satisfactorily effected by boring a hole deeply 
 into a certain proportional number of the pieces to be 
 sampled. 
 
 Where such work is only exceptional, an ordinary 
 ratchet hand-drill will answer, but in most cases a half- 
 inch drill run by machinery is employed. 
 
 The chips and drillings are still further subdivided by 
 scissors, and, as even then it is difficult to obtain an abso- 
 lutely perfect mixture, it is best to weigh out and dissolve 
 a much larger amount than is usually taken for assay, 
 
14 PREPARATION OF THE SAMPLE. 
 
 taking a certain proportion of the thoroughly mixed solu- 
 tion for the final determination. 
 
 The ore must always be reduced to a powder, more or 
 less fine, according to the nature of the chemical operation 
 or assay to which it is to be subjected. This division is 
 effected by means of the anvil, hammer, pestle and mortar, 
 sieve, method of elutriation, or other means generally in 
 use for the preparation of any fine powder. The actual 
 process to be adopted must vary according to the nature 
 of the different bodies under examination. In some cases 
 simple crushing is sufficient ; in others the ore will have 
 to be pounded in a mortar ; whilst occasionally it is neces- 
 sary to reduce it to the very highest degree of fineness by 
 elutriation. 
 
 There are other operations us strictly mechanical as 
 are the above, viz. washing, dressing, and vanning a 
 sample of ore, the end and aim of which is to separate, 
 in a suitable vessel, by means of water and difference of 
 specific gravity, the earthy or useless and, in some cases, 
 objectionable portion from the heavier metallic and valu- 
 able portion. This operation is almost always employed 
 on the larger scale in dressing ores for the smelter. 
 
 The tools and materials employed in preparing the 
 sample in the assay laboratory are the anvil (and stand), 
 vice, hammer, files, cold chisel, shears, pestle and mortar, 
 *t eel-crushing mortar, sieve, &c. 
 
 THE ANVIL (fig. 1). The anvil is most useful in size 
 when it weighs about 28 Ibs. ; but one of 14 Ibs. will 
 suffice. The anvil recommended is of the shape usually 
 employed by the blacksmith. 
 
 The anvil-stand is constructed of stout wood, about 
 two inches in thickness, and forms a cube of about two 
 feet square. It contains three or four drawers, which 
 serve to hold the hammers, cold chisel, shears, files, &c., 
 which are required in an assay office. In the centre the 
 anvil is fixed, and in one corner: a vice may be also 
 secured. 
 
 In general the anvil and hammer are employed for the 
 purpose of breaking a small fragment, from a mass of ore 
 
PREPARATION OF THE SAMPLE. 
 
 15 
 
 for examination, or ascertaining whether the button or 
 prill of metal produced in an assay be malleable or other- 
 
 FlG. 1. 
 
 wise. The anvil is also exceedingly useful as a support 
 for a crucible while breaking it to extract the metallic or 
 other valuable contents. 
 
 THE HAMMERS (figs. 2 and 3), of which two are requisite, 
 ought to have one end flat and FIG. 2. FIG. 3. 
 square and the other pick- or 
 wedge-shaped. The horizontal 
 wedge end of fig. 2 is useful for 
 breaking open crucibles and in 
 detaching small fragments from 
 a specimen of ore. The flat end 
 serves for ascertaining the mal- 
 leability of buttons of metal. 
 This hammer should weigh about 
 1 Ib. The larger hammer, fig. 3, 
 should weigh about 4 Ibs., and is 
 employed for breaking coke suffi- 
 ciently fine for the use of the fur- 
 nace, and detaching fragments 
 from refractory minerals, in both of which cases either 
 
16 PREPARATION OF THE SAMPLE. 
 
 end may be employed, as may seem most serviceable to 
 the operator. The flat end of this hammer is also used 
 for driving a cold chisel in separating masses of gold, 
 silver, copper, lead, &c., for assay. This hammer has a 
 vertical pick or wedge end. 
 
 Very hard and stony materials which have to be broken 
 on the anvil (and all such ought to be so treated) scatter 
 many fragments, to the certain loss of a proportion of the 
 substance, and the probable injury of the operator ; this 
 can be prevented by wrapping the mineral in a piece of 
 stout brown paper, or if necessary in several folds. The 
 fracture can then be safely attempted. 
 
 This latter precaution must be specially taken in frac- 
 turing gold quartz, or hard rock containing metallic silver, 
 
 FIG. 4. FIG. 5. 
 
 as the loss of a very minute quantity of metal would 
 involve a considerable error in the result afforded by the 
 
 assay. 
 
 All minerals, unless very friable, must be reduced to a 
 moderate size say that of a walnut by means of the 
 anvil and hammer, before pulverisation ; otherwise, if the 
 reduction be attempted in a mortar, it is nearly certain to 
 be injured ; moreover, the operator will find his labours 
 much abridged by using the anvil for this purpose. 
 
 The anvil can also be made very serviceable in repoint- 
 ing worn or burnt-out tongs &c. It need scarcely be 
 added that it must be placed as far as possible away from 
 bottles or other frangible articles, otherwise accidents may 
 occur by the forcible projection of fragments of crucibles, 
 
 stones, &c. 
 
 THE COLD CHISEL (fig. 4) is employed for cutting off me- 
 tallic masses for assay. It should be five or six inches long 
 and about half an inch wide, which is the best size for 
 general use. However, for some purposes, as cutting copper 
 
PREPARATION OF THE SAMPLE. 17 
 
 and other very tough metals, it is convenient to have a 
 chisel only a quarter of an inch wide, as these metals are 
 so much more difficult to cut, and the small chisel meets 
 with the least resistance. 
 
 Small shears (fig. 5) are also exceedingly useful in 
 cutting off pieces of sheet metal, such as lead, for cupel- 
 lation, scorification, &c. 
 
 THE PESTLE AND MOKTAR. Mortars are made of various 
 materials, as cast-iron, bronze, porcelain, agate, &c. ; the 
 assay er requires one of cast-iron, one of porcelain, and 
 one of agate. 
 
 The iron mortar (fig. 6) ought to be of the capacity of 
 from three to four pints ; the porcelain (Wedgwood ware) 
 (fig. 7) may contain about two pints. The ease with which 
 
 FIG. 6. FIG. 7. 
 
 a mortar may be used depends much upon its form, and 
 opinion is greatly divided on the subject. Faraday* says 
 that the pestle should be strong, and the size of its upper 
 part sufficient to allow of its being grasped firmly in the 
 hand, and below to permit a considerable grinding surface 
 to come in contact with the mortar. Its diameter in the 
 lower part may be about one third or one fourth of the 
 upper diameter of the mortar. The curve at the bottom 
 should be of shorter radius than the curve of the mortar, 
 that it may not touch the mortar in more than one part, 
 whilst at the same time the interval around may gradually 
 increase, though not too rapidly, towards the upper part 
 of the pestle. 
 
 The bottoms of all mortars ought to be of considerable 
 
 * Chemical Manipulation, p. 149. 
 
18 PREPARATION OP THE SAMPLE. 
 
 thickness, in order to withstand the smart blows they will 
 occasionally have to receive. 
 
 Berzelius recommended a mass of pumice-stone for 
 cleansing porcelain mortars. It is used with water as a 
 pestle, and in course of time will be worn to the shape of 
 the mortar ; its action will then be more speedy. 
 
 Iron mortars can be best cleaned by friction with a 
 little fine sharp sand, if washing be not sufficient to re- 
 move the adhering substance. Great care must be taken 
 to dry mortars perfectly, especially those of iron, other- 
 wise they will become rusted, and the rust will contaminate 
 the substances pulverised in them. 
 
 The iron mortar is principally of use in the reduction 
 of the masses of mineral (broken on the anvil, as before 
 described) to a state of coarse powder, in order to render 
 the substance more readily capable of pulverisation, strictly 
 so called. In the use of the iron mortar, all friction with 
 the pestle ought to be avoided, and the body within it 
 must be struck repeatedly and lightly, in a vertical direc- 
 tion, taking care to strike the large pieces, so that all may 
 be equally reduced. This can be carried on until the 
 whole is about the size of fine sand. It is transferred to 
 the porcelain mortar, where direct blows must be carefully 
 avoided. 
 
 The process is now carried on somewhat differently ; 
 the pestle is to be pressed with a moderate force, and a 
 circular motion given to it, taking care every now and 
 then to lessen and then to enlarge the circles so as to pass 
 over the whole grinding surface of the mortar, and insure 
 the pulverisation of the mass of mineral submitted to 
 operation. In general, the finer the state of division to 
 which a mineral is reduced, the more accurate and ex- 
 peditious will be its assay ; and in preparing a mineral for 
 assay by the wet method, no labour ought to be spared 
 on this point. Pulverisation is rendered much easier by 
 operating on a small quantity at once, and removing it 
 very often from the sides and bottom of the mortar by 
 means of a spatula. The quantity operated on at one 
 time must be regulated by the hardness and friability of 
 
PREPARATION OF THE SAMPLE. 19 
 
 the substance whose pulverisation is to be effected. The 
 harder it is, the less must be taken, and vice versa. 
 
 In the use of the iron mortar fragments are occasionally 
 projected. This may be prevented by covering the upper 
 part of the mortar with a cloth. This applies also to the 
 porcelain mortar, for the dust of some minerals has a 
 disagreeable taste and smell. Indeed, in some cases the 
 ambient powder is highly deleterious, as in the pulverisa- 
 tion of arsenical nickel, cobalt, and other ores. Here the 
 simple cloth is not a sufficient protection ; it should be 
 slightly damped with water, and tightly tied round the 
 mortar, and firmly held round the pestle, when nothing 
 can escape. 
 
 Some minerals can be pulverised with greater ease if 
 they are ignited and suddenly quenched in cold water. 
 Amongst them may be named flint, and many other sili- 
 ceous matters, as gold quartz. In the pulverisation of 
 charcoal for assays, it will be found useful to heat it, as 
 hot charcoal is more readily pulverised than cold. 
 
 In some instances the powder obtained in the iron or 
 porcelain mortar is not fine enough ; recourse should then 
 be had to the agate mortar, in which the mineral, in as 
 fine a state of division as the larger mortars will give it, is 
 ground in small portions at 
 a time, until it is reduced to 
 an impalpable powder. 
 
 When small specimens or 
 rare minerals are being ope- 
 rated upon, if it is especially 
 desirable to avoid loss, it is 
 advisable to use a steel mor- 
 tar (fig. 8) for the prepara- 
 tory reduction of the mineral 
 to coarse powder. A B, C D, 
 and E F represent the three 
 component parts of the mor- 
 tar ; these may be readily 
 taken asunder. The substance 
 to be crushed (having, if practicable, first been broken into 
 
 c 2 
 
L>0 PREPARATION OF THE SAMPLE. 
 
 small pieces) is placed in the cylindrical chamber E F; 
 the steel cylinder, which fits somewhat loosely into the 
 chamber, serves as a pestle. The mortar is placed upon 
 a solid support, and perpendicular blows are repeatedly 
 struck upon the pestle with a hammer, until the object in 
 view is attained. (Fresenius.) 
 
 In the selection of agate mortars, they must be examined 
 to see that they have no palpable flaws in them ; very 
 slight cracks, however, that cannot be felt, do not render 
 the mortar useless, although they increase the danger of 
 its destruction by a chance blow. 
 
 THE SIEVE. The operation of sifting is employed when 
 a very fine powder is required, or when a powder of 
 uniform size is needed. Sieves of various materials and 
 different degrees of fineness are necessary. The larger 
 sieve, for preparing coke for the blast furnace, is made of 
 stout iron wire, and must have its meshes from 1 inch to 
 1J inch square. The fine coke, which is sifted from that 
 which is the proper size for the blast furnace, may be 
 mixed with that of ordinary size, and employed economi- 
 cally in the muffle furnace. For the preparation of 
 minerals a set of three sieves should be provided, each 
 one finer than the other. The coarsest may contain 40 
 holes to the linear inch, the finer or medium sieve 60, and 
 the finest from 80 to 100. The coarsest sieve is used for 
 preparing galena for assay ; the medium for copper, tin, 
 iron, and other like ores ; and the finest for gold and 
 silver ores, or for preparing any substance for the wet 
 assay, as, in the latter case, the finer the state of division 
 the substance attains, the more rapid will be its solution 
 or decomposition by the liquid agents employed. 
 
 The sieve fig. 9 is made of wood, over which is strained 
 in the ordinary manner brass wire-gauze of the necessary 
 degree of fineness. When in use, j5, fig. 10, is fitted into 
 the lower part of A (same figure). This contrivance pre- 
 vents all loss of the fine powder. If the matter to be 
 sifted be offensive or deleterious to the operator, a sieve 
 termed the drum or box-sieve may be employed (see fig. 
 10), where C represents a cover fitting over the sieve. If 
 
SIFTING. 21 
 
 small, this may be used in the ordinary way ; but if large, 
 its method of use is rather peculiar, and requires some 
 practice to fully develop its powers. One side of the 
 under edge must be FIG. 9. FIG. 10. 
 
 held by one or both 
 hands according to its 
 size, whilst the other 
 rests on a table or a 
 bench. A semicircu- 
 lar oscillating motion 
 must now be com- 
 municated to it by 
 moving the hands 
 up and down at the 
 same time that they 
 are being alternately brought into approximation with the 
 sides of the operator. 
 
 In cases of necessity, a sieve may be readily extem- 
 porised. Place the powder to be sifted in a piece of fine 
 lawn or muslin, according to the fineness required, tie it 
 up loosely, and shake or tap the powder, with, its muslin 
 or other envelope, on a sheet of paper, and the sifting will 
 be rapidly and easily accomplished. 
 
 The sieve is also extremely serviceable in the separation 
 of some ores from their gangues or vein-stones, especially 
 if the latter be stony and hard. This point must be par- 
 ticularly noted, as it is the cause of much variance between 
 the results of different assayers ; for instance, part of the 
 same sample of ore might be sent to two assayers, and the 
 produce made by one would be 8^ per cent., and that by 
 the other 9 or 9^, or, in some cases, even more. This dis- 
 crepancy generally arises from the cause above mentioned. 
 In the one case the workman has rejected part of the 
 hard gangue, and so rendered the residue richer ; whilst 
 in the other he has pulverised the whole, making the 
 produce less, but giving more accurately the amount of 
 .metal in the substance submitted to assay. 
 
 A knowledge of this fact is also very useful from 
 another point of view. Suppose it were wished to separate 
 
'2*2 SIFTING. 
 
 in a speedy manner, as perfectly as possible, any friable 
 mineral, such as galena or copper pyrites, from its matrix 
 by mechanical means, it might be accomplished by the 
 use of the sieve, as follows : Place a small quantity of 
 the mineral in an iron mortar, and strike, repeatedly, 
 slight vertical blows. When it is tolerably reduced, sift 
 it, and it will be found that what passes through is nearly 
 pure mineral, with only a small quantity of matrix ; repeat 
 the pounding and sifting operations, until, after a few 
 repetitions, that which remains in the sieve is nearly pure 
 gangue. 
 
 Native metals, as gold, silver, and copper, are also* 
 partially separated after the manner above described. 
 The fine particles of metal, during the process of pounding 
 and trituration, become flattened, and cannot pass through 
 the sieve, whilst the more brittle portions pass through 
 and are separated. 
 
 ELUTRIATION. This process can only be employed for 
 those bodies which are not acted on by water ; and it must 
 be remembered that many substances which are usually 
 considered to be insoluble in water are, when in a very 
 finely divided state, acted upon to a greater or less extent. 
 The operation is thus effected : The substance is reduced 
 to the finest possible state of division by any of the fore- 
 going processes ; it is then shaken up with a quantity of 
 water in a glass or other vessel, After a few moments' 
 repose, the supernatant liquid, retaining in suspension the 
 finer particles of the pulverised substance, is poured off, 
 and the grosser parts, which have fallen to the bottom of 
 the vessel, are repulverised, and again treated with water. 
 By repeating these processes a powder of any required 
 degree of fineness may be obtained. 
 
 It is seldom, however, that a substance is required for 
 assay by the dry way, in such a minute state of division. 
 In the humid or wet method it is occasionally very useful. 
 If the supernatant water is roughly decanted off, where 
 the powder to be elutriated is light, the least disturbance 
 of the vessel containing it occasions the distribution of the 
 portion which has settled, throughout the liquid, and the 
 
ELUTRIATION. 23 
 
 consequent mixture of fine and coarse particles. This 
 can be avoided by the employment of the syphon. The 
 operation is then thus conducted : The syphon is filled 
 with water, and the shorter end placed in the liquid whose 
 transversion is to be effected : the forefinger of the right 
 hand, which, during this time, has been applied to the 
 longer end of the instrument, is now removed, when the 
 water will flow out until it is level with the immersed 
 end of the syphon. Fresh water can then be added, the 
 powder stirred up again, and the operation of decantation 
 by the syphon carried on as long as requisite. 
 
 WASHING, DRESSING, OR VANNING. This operation is ex- 
 ceedingly useful for discovering the approximate quantity 
 of pure ore, such as galena, copper pyrites, oxide of tin, 
 native gold or silver, in any sample of earthy matter or 
 ore in which it may be disseminated. 
 
 The theory of the operation about to be described is 
 easily understood. Bodies left to the action of gravity in 
 a liquid, in a state of rest, experience a resistance to their 
 descent which is proportionate to their surface, whatever 
 may be their volume and density. Hence it follows, firstly, 
 that of equal volumes the heaviest fall most rapidly; 
 secondly, that of equal densities those having the largest 
 size move with the greatest speed ; for in particles of un- 
 equal size and like form the weight is proportional to the 
 cube of the dimensions, whilst the surface is only propor- 
 tional to the square of these dimensions ; hence in small 
 particles the surface is greater in relation to the weight 
 than in the large particles. Thirdly, of equal densities and 
 volumes, particles offering the largest surface (those which 
 are scaly and laminated, for example) undergo more re- 
 sistance in their motion than those which, approaching 
 the spherical form, have less surface. The adhesion of 
 the liquid to the particles of bodies held in suspension is 
 also an obstacle to their subsidence. This force, like the 
 dynamic resistance, is proportional to the surface and in- 
 dependent of the mass or volume ; whence it follows that, 
 in a fluid in motion, of bodies having equal volumes, the 
 least dense acquire the greatest rapidity of movement, and 
 
24 DRESSING OR VANNING. 
 
 are deposited at the greatest distance from the point of 
 departure ; whilst with equal densities the smallest grains 
 are carried farthest ; and lastly, with equal densities and 
 volumes, the particles exposing most surface traverse the 
 greatest space. 
 
 It is, therefore, evident that the most advantageous 
 condition for separating, by washing, two substances of 
 unequal specific gravity or density is that the heavier 
 shall be in larger grains than the lighter ; this unfortu- 
 nately, however, is a condition that can very seldom be 
 fulfilled, as the heaviest substances are those metallic 
 minerals whose frangibility is nearly always greater than 
 the earthy matters accompanying them as gangues. This 
 being the case, it is very important so to arrange that the 
 fragments of the various mixed substances shall be nearly 
 of the same size. This may be effected by very frequently 
 sifting the mineral during the process of pulverisation, 
 reducing it also more by blows than by grinding, so as 
 to get as little fine powder as possible, as that is nearly 
 certain to be washed away during the process. 
 
 The operation of washing or vanning may be performed 
 by one of two methods. In the first, a small stream of 
 running water is employed ; in the second, water is added to 
 the substance to be washed, and poured off as necessary. 
 
 In the first process, a vessel somewhat resembling a 
 banker's gold scoop (but longer in proportion) is employed ; 
 the mineral to be washed is placed in the upper part, and a 
 small quantity of water added, with which the mineral is 
 thoroughly and carefully moistened, and mixed with the 
 fingers. The scoop must then be so inclined that a fine 
 stream of water from any convenient source (say a tap) may 
 fall just above the upper part of the mixture of mineral 
 and water ; then, firmly holding the larger and upper end 
 of the scoop with the left hand, and sustaining the lower 
 part with the right, it is shaken frequently in the direction 
 of its longitudinal axis. At each shake all the particles 
 in the scoop are so agitated that they become suspended 
 in the water, and the current of liquid running from the 
 tap into the scoop moves them all in its own direction ; 
 
DRESSING OR VANNING. 25 
 
 but they are deposited at different distances from the 
 point at which the water enters, the heaviest being carried 
 through but a very small space. It is now soon seen that 
 the mineral assumes a heterogeneous surface ; at the upper 
 part, the heavy portions are seen nearly pure ; the light 
 substances, on the other hand, are nearly without mixture 
 at the lower end, and in the intermediate part the heaviest 
 portion of the mixture is nearest the upper end. If the 
 washed matter were now to be divided into horizontal 
 layers, the heaviest matter would be found at the bottom, 
 and the lightest on the surface. Things being in this 
 state, the scoop must be made to oscillate on its axis, so 
 that the latter remain immovable, and in a slightly in- 
 clined position. In this manner, the layer of water 
 running over the surface of the mineral agitates that part 
 only, and carries off all light substances there deposited in 
 the previous operation. When necessary, these matters 
 may be removed by the finger, and made to run into a 
 vessel placed below the scoop, in which all the water and 
 matters carried off are received. This operation, however, 
 must not be hurriedly performed, so as to mix the parts 
 already separated : each layer must be removed separately, 
 commencing with the upper one. This being done, the 
 scoop must be alternately kept in motion by shakings, as 
 at first, and then on its axis, and the washing off of the 
 finer particles renewed, and so on until the separation is 
 effected as far as may be judged necessary. 
 
 At the commencement of the operation, the water 
 carries out of the scoop the lightest particles, as organic 
 matter, clay, &c, ; at a little later period the water carries 
 with it a small but definite quantity of the heavier 
 portion, the proportion of which increases as the operation 
 proceeds, until at last the greatest possible care is required. 
 It is always better to rewash the latter portion which 
 passes off from the scoop ; hence the necessity of allowing 
 all the wash-water passing from it to collect in a vessel 
 placed for that purpose. 
 
 In the second method of washing, a tin, zinc, or wooden 
 pan is employed. It should be circular, one or two feet in 
 
20 DRESSING OR VANNING. 
 
 diameter and three or four inches deep ; the sides should 
 descend in a conical manner, so that the bottom is not more 
 than four inches in diameter, and the angle between it and 
 the sides as sharp as possible. 
 
 The substance to be examined is placed in the washing- 
 dish, the latter filled with water, and the mineral well mixed 
 with it until perfectly moistened as before. After a moment 
 or so the muddy water is poured off, and the operation 
 repeated until the water passes off clear. When this 
 happens, only so much water must be placed in the pan as 
 will leave a slight layer on the mineral. Now, by holding 
 the pan in one hand, and shaking it with the other, the 
 greater part of the heavy mineral, gold or otherwise, will 
 fall below the sand. If now the pan be inclined towards 
 the hand which is shaking it, the lighter portions, even if 
 tolerably large, will flow off with the water, leaving the 
 heavier matters in the angle, from which, with ordinary 
 care and a little practice, it is difficult to disturb them* 
 If there be a large quantity of earthy matter, this may be 
 (after sufficient shaking) removed by the finger, as in the 
 first-described process. By careful repetitions of these 
 processes, the whole, or nearly the whole, of the sandy 
 and earthy matters may be removed, and the gold or 
 other mineral left nearly pure. This is the plan employed 
 in prospecting for gold, diamonds, and other gems, and in 
 some cases for their commercial extraction. 
 
 In Cornwall and other mining counties this operation 
 is very cleverly and carefully performed on the miner's 
 common shovel, and the richness of any particular sample 
 of either tin, lead, or copper is thereby determined with a 
 very near approach to accuracy. 
 
 THE BALANCE. 
 
 OPERATION OF WEIGHING. At least three balances will 
 be required in a laboratory where general assays are per- 
 formed. The first must be capable of carrying three or 
 four pounds in each pan, and must turn with a quarter of 
 a grain. This may be of the form of the bankers' or 
 bullion balance (fig. 11), and may be employed in weigh- 
 
WEIGHING. 
 
 ing samples of gold quartz or silver ore containing metallic 
 grains capable of being separated by the sieve (see p. 21) ; 
 the second (fig. 12), or rough assay balance, is similar to 
 
 FIG. 11. 
 
28 WEIGHING. 
 
 the apothecary's scales ; it should take 1,000 grains in 
 each pan, and turn with one tenth of a grain. This serves 
 for weighing samples of ore and fluxes for assay, and for 
 determining the weight of buttons or prills of lead, tin, 
 iron, copper, &c., obtained in an assay. 
 
 The third and most delicate^ or true, assay balance (fig. 
 13) should carry about 1,000 grains; must turn distinctly 
 and accurately with the -roW tn of a grain. This is 
 
 employed in the assay of gold and silver, bullion, and in 
 the assay of minerals containing gold and silver ; also for 
 general analytical purposes. The first two balances may 
 be used, with ordinary care, by any one ; but the third 
 balance, in its use and adjustment so as to maintain its 
 extreme accuracy, requires some particular instructions, 
 which necessarily involve the principle of the balance. 
 These have been so admirably given by Faraday, in his 
 * Chemical Manipulations,' that we can do no better than 
 transcribe them : 
 
 4 The theory of this balance is so simple that the tests 
 of its accuracy will be easily understood and as easily 
 practised. It may be considered as a uniform inflexible 
 
THE BALANCE. 29 
 
 lever, supported horizontally at the centre of gravity, and 
 supporting weights at equal distances from the centre by 
 points in the same horizontal line with the centre of 
 gravity. If the weights be equal the one will counter- 
 poise the other ; if not, the heavier will preponderate. 
 In the balance, as usually constructed, there are certain 
 departures from the theory as above expressed some 
 from the impossibility of execution, and others in conse- 
 quence of their practical utility ; and a good balance may 
 be said to consist essentially of a beam made as light as is 
 consistent with that inflexibility which it ought to possess, 
 divided into two arms of equal weight and length by a 
 line of support or axis, and also terminated at the end of 
 each arm by a line of support or axis, intended to sustain 
 the pans. These three lines of support should be exactly 
 parallel to each other in the same horizontal plane, and 
 correctly perpendicular to the length of the beam ; and 
 the plane in which they lie should be raised more or less 
 above the centre of gravity of the beam, so that the latter 
 should be exactly under the middle line of suspension. 
 It will be unnecessary in this place to speak of the coarse 
 faults which occur in the ordinary scales these will be 
 easily understood ; and from what has to be stated of the 
 examination of the most delicate instrument, the impos- 
 sibility of avoiding them without incurring an expense 
 inconsistent with their ordinary use will be as readily 
 comprehended.' 
 
 Two principal things have to be attended to in the 
 selection of a balance its accuracy and its delicacy. The 
 accuracy depends upon the following conditions : 
 
 1. The arms should be equal to each other in length. The 
 length of each is accurately the distance from the middle 
 to the distant knife-edge, all the edges being considered 
 parallel to each other, and in the same plane. The two 
 arms should accord perfectly in this respect. This equality 
 may be ascertained in two or three ways. Suppose the 
 balance with its pans to vibrate freely, and rest in a 
 horizontal position, and that after changing the pans from 
 one end to the other the balance again takes its horizontal 
 
30 WEIGHING. 
 
 state of rest in such a case an almost certain proof is 
 obtained of equality in length of the arms. They may, 
 however, be equal, and yet this change of the pans from 
 end to end may occasion a disturbance of equilibrium, 
 because of the unequal distribution of weight on the beam 
 and pans ; but to insure an accurate test, restore the pans, 
 and consequently the equilibrium, to the first state : put 
 equal, or at least counterpoising, weights into the pans, 
 loading the balance moderately, and then change the 
 weights from one pan to the other, and again observe 
 whether the equilibrium is maintained ; if so, the length 
 of the arms is equal. 
 
 Equality of weight is not so necessary a condition, 
 although this should be obtained as accurately as possible. 
 One arm with its pan may be considerably heavier than 
 the other, but from the disposition of the weight in the 
 lighter arm towards the extremity, or in the heavier 
 towards the middle of the beam, the equilibrium may be 
 perfect, and therefore no inaccuracy be caused thereby in 
 the use of the balance. Instruments are usually sent home 
 In equilibrium, and require no further examination as to 
 this particular point than to ascertain that they really are 
 in adjustment, and that after vibrating freely they take a 
 horizontal position. 
 
 2. The beam must be of such a form and strength that 
 it will not bend when loaded with the greatest weight the 
 balance is intended to carry. All well-made modern 
 balances are sufficiently rigid in this respect, and may 
 be safely trusted to carry their full weight without 
 flexure of the beam. The beam should also be as light as 
 practicable. 
 
 3. The knife-edges supporting the pans, and the centre 
 one on which the beam vibrates, must be accurately in the 
 same line. 
 
 The delicacy of a balance likewise depends upon 
 several conditions. 
 
 The centre of gravity must be very little below the 
 fulcrum. If it be considerably depressed, then, upon 
 trying the oscillations of the balance by giving it a little 
 
THE BALANCE. 31 
 
 motion, they will be found to be quick, and the beam 
 will soon take its ultimate state of rest ; and if weights be 
 added to one side, so as to make it vibrate, or to bring it 
 to a certain permanent state of inclination, the quantity 
 required will be found to be comparatively considerable. 
 As the centre of gravity is raised the oscillations are 
 slower, but producible by a much smaller impulse ; the 
 beam is a longer time before it attains a state of rest, and 
 it turns with a smaller quantity. 
 
 If, however, the centre of gravity coincides with the 
 fulcrum or centre of oscillation, then the balance is said 
 to se^ that is, the smallest possible weight will turn the 
 beam ; the oscillations no longer exist, but one side or 
 the other preponderates with the slightest force, and the 
 valuable indication which is furnished by the extent and 
 velocity of the vibrations is lost. 
 
 The case in which the centre of gravity is above the 
 fulcrum rarely if ever occurs. Such a balance, when 
 equally weighted, would set on the one side or the other ; 
 that side which was in the slightest degree lower tending 
 to descend still further, until obstructed by interposing 
 obstacles. 
 
 In balances intended to carry large quantities (as in 
 the balance for weighing gold quartz, &c.) it is necessary 
 to place the centre of gravity lower than in those for 
 minute quantities, that they may vibrate regularly and 
 readily. This is one cause why they are inferior in deli- 
 cacy, for, as a consequence of the arrangement, they will 
 not turn except with a larger weight. 
 
 Balances are also liable to set when overloaded. Thus, 
 if a balance be equally weighted in each pan, but over- 
 loaded, it will, if placed exactly horizontal, remain so, .but 
 the slightest impulse or depression on one side destroys 
 the equilibrium ; the lower side continues to descend with 
 an accelerated force, and ultimately remains down, being 
 to all appearance heavier than the other. Generally 
 speaking, the more delicate a balance the sooner this 
 effect takes place ; this is one limit to the weight it can 
 properly carry. 
 
32 WEIGHING. 
 
 The vibrations of a balance vary with the quantity of 
 matter with which it is loaded : the more the weight in 
 the pans, the slower the vibrations. These should be 
 observed, and the appearances retained in the mind, in 
 consequence of the useful indications they afford in 
 weighing. A certain amplitude and velocity of vibration 
 would indicate to a person used to the instrument nearly 
 the weight required to produce equilibrium ; but the same 
 extent and velocity, with a weight much larger or smaller, 
 would not be occasioned by an equal deficiency or redund- 
 ancy of weight, as in the former case. 
 
 The weight also required to effect a certain inclination 
 of the beam, or to turn it, should be known, both when it 
 is slightly and when it is heavily loaded. Thus, if the 
 instrument turns with T ^Vo tn of a grain, with 1,000 grains 
 in each pan, or with T o~o-g-^- o^h of the weight it carries, it 
 may be considered perfect. 
 
 The friction of the knife-edges must be as slight as 
 possible. 
 
 Most of the faults in the working of a balance, if 
 ordinarily well made, depend upon imperfections in the 
 middle knife-edge and the planes upon which it rests. 
 
 The edge is made either of agate or ste.el, preferably 
 the former, and should be formed out of one piece, and 
 finished at once, every part of the edge being ground on 
 the same flat surface at the same time. In this way the 
 existence of the two extreme or bearing parts of the edge 
 in one line is insured ; but when the two parts which bear 
 upon the planes are formed separately on the different 
 ends of a piece of agate or steel, or, what is worse, when 
 they are formed on separate pieces, and then fixed one on 
 each side the beam, it is scarcely possible they should be 
 in the same line : and if not, the beam cannot be correct. 
 These knife-edges usually rest on planes, or else in curves. 
 The planes should be perfectly flat and horizontal, and 
 exactly at the same height ; the curves should be of equal 
 height, and their axes in the same line. If they are so, 
 ancl the knife-edge is perfect, then the suspension will be 
 accurately on the line of the edge, and reversing the beam 
 
THE BALANCE. 33 
 
 will produce no change. The balance must always be 
 kept perfectly level by means of the three screws on 
 which it stands, and adjusted by the spirit-level or plum- 
 line with which it is furnished. 
 
 The balance should be kept in a well-lighted dry room, 
 quite away from acid or other vapours. The case should 
 be kept closed as much as possible, and a glass vessel full 
 of lumps of good quick-lime should be kept in it. When 
 the lime falls to powder it should be renewed. 
 
 In order to test the accuracy and delicacy of a balance, 
 remove the pans and their end supports, and notice how 
 the beam oscillates. When it has been found to oscillate 
 with regularity, and gradually to attain a horizontal posi- 
 tion of rest, it should be reversed -that is, taken up and 
 turned half-way round, so as to make that which before 
 pointed to the right now point to the left. The beam 
 should then again be made to oscillate, and if it perform 
 regularly as before, finally resting in a horizontal position, 
 it has stood a severe test, and promises well. Then re- 
 place the pans and repeat the tests, noticing the time 
 required for each oscillation. When the pans are hung 
 upon the beam, the balance should of course remain 
 horizontal. They should be tried by changing, then by 
 reversing the beam, and afterwards by changing the pans 
 again. The pans are best suspended by very thin platinum 
 wire, so as to avoid hygrometrical influence upon them. 
 
 Afterwards load the balance with the full weight it 
 is intended to carry say 1,000 grains in each pan, and 
 notice if the indications are as rapid upon adding or 
 subtracting the smallest weight as they were when the 
 pans were empty. 
 
 Tests of this kind are quite sufficient for the purpose 
 of the assayer, who, having ascertained that his balance, 
 whether slightly or fully laden, vibrates freely, turns deli- 
 cately, and has not its indications altered by reversing the 
 beam or changing counterpoising weights, may be perfectly 
 satisfied with it. 
 
 The irregularities which may be discovered by these 
 tests are best corrected by a workman ; but as in all the 
 
 D 
 
34 WEIGHING. 
 
 best balances now made adjusting screws for these pur- 
 poses are provided, it has been thought advisable to 
 introduce here such matter as, after careful perusal, will 
 enable every one to adjust and examine his balance 
 properly ; so that, in the absence of a skilled workman, it 
 may without much danger be put into working order by the 
 assay er himself, if accidentally damaged by rough treatment. 
 
 THE WEIGHTS. Various kinds of weights are necessary 
 for the different balances required by the assay er. For 
 the larger balance, Troy-weights from 4 Ibs. to ^ grain 
 will be requisite ; for the second size, weights from 1,000 
 grains to y^th part of a grain ; and for the assay balance, 
 weights from 1,000 grains to ToVo tn f a g ram - 
 
 The best material adapted for weights is unquestion- 
 ably platinum. This is, however, too expensive for its 
 general adoption, and therefore brass weights are almost 
 invariably employed down to the ten- or twenty-grain 
 weight, the smaller ones only being of platinum. On the 
 Continent weights are generally made of silver, and if of 
 brass are electro-gilt. For the smallest weights of all 
 (those below 0*10 grain) aluminium is often used, its 
 lightness, and consequently greater bulk, enabling these 
 small weights to be made considerably larger than if they 
 were of platinum. The riders are generally of silver-gilt 
 wire. The slight tarnish which gradually forms on brass 
 weights may be disregarded until it becomes thick. 
 Weights ought never to be touched with the fingers, and 
 should, when not in use, be kept tightly fastened in their 
 box, away from all acid fumes. The most convenient series 
 in which to have the weights is 600, 300, 200, 100, 60, 30, 
 20, 10, 6, 3, 2, 1, -6, -3, -2, -1, &c. This is preferable 
 to the series formerly employed, as it admits of the use of 
 a less number of weights to arrive at any required amount. 
 
 According to Deville and Mascart receptacles lined 
 with velvet are not adapted for the preservation of weights, 
 as dust is deposited in the velvet and acts upon the weights 
 when taken out or put in. Small boxes of ivory or smooth 
 wood are preferable. 
 
 Peculiar weights are necessary for the assay of gold 
 and silver bullion in England (with the exception of assays 
 
GOLD AND SILVER ASSAY POUNDS. 
 
 35 
 
 for the Bank of England ; see Gold assay), gold being 
 reported in carats, grains, and eighths, and silver in ozs. 
 and dwts. The most convenient quantity of either of 
 the precious metals for assay is 12 grains. The quantity 
 taken, however, is of no very great consequence ; but 
 whatever its real weight, it is denominated in England 
 the assay ' pound' This assay ' pound ' is then subdivided 
 into aliquot parts, but differing according to the metal. 
 The silver assay ' pound ' is subdivided, as the real Troy 
 pound, into 12 ounces, each ounce into 20 pennyweights, 
 and these again into halves (the lowest report for silver), 
 so that there are 480 different reports for silver, and 
 therefore each nominal half-pennyweight weighs ^th part 
 of a Troy grain, when the ' pound ' is twelve grains. 
 
 Assay Weights for Silver. 
 
 Assay 
 grains. 
 
 12 
 11 
 
 6 . 
 3 
 2 
 1 
 
 0-500 
 0-250 
 0-150 
 0-100 
 0-050 
 0-025 
 
 The gold assay 'pound is subdivided into 24 carats, 
 each carat in to 4 assay grains, and each grain into eighths, 
 so that there are 768 reports for gold; and the assay 
 * pound ' weighing 12 Troy grains, the lowest report, or -|th 
 assay grain, equals -g^th Troy grain ; thus 
 
 Assay Weights for Gold. 
 
 Silver 
 ozs. dwts. grs. 
 
 12 
 
 
 11 
 
 
 600 
 
 
 300 
 
 
 200 
 
 
 100 
 
 
 10 
 
 
 050 
 
 
 030 
 
 
 020 
 
 
 010 
 
 
 12 
 
 
 ( 
 
 carats. 
 
 24 
 22 
 12 
 
 Joid 
 grs. 
 
 
 
 
 
 o 
 
 eighths. 
 
 . 
 . 
 
 o . . 
 o 
 
 Assay 
 grains. 
 
 12 
 11 
 . . . .6 
 3 
 
 3 
 
 o 
 
 o 
 
 Iffths 
 
 2 
 1 
 
 
 o 
 
 
 
 2 
 1 
 
 . 
 . 
 . . 
 
 o 
 
 '. '. '. '. ffths 
 
 
 
 
 
 6 . 
 
 - 6 -ths 
 
 o 
 
 o 
 
 3 
 
 e?ths 
 
 
 
 
 
 
 
 2 . 
 
 1 
 
 
 D 2 
 
36 THE WEIGHTS. 
 
 In cases where the very smallest weights have to be 
 employed, great care must be taken in seizing them with 
 the forceps, as they are apt to spring away and be lost. 
 In the assay balance (fig. 13) the use of weights less than 
 -j^thofa grain is avoided by a very ingenious contrivance. 
 Each side of the beam is equally divided into ten parts, 
 and over the beam on each side is placed a sliding rod, 
 as represented in the figure. The object of these rods is 
 to carry, in the direction of the beam, the small bent piece 
 of wire (letter c, fig. 13) called a rider, which serves in 
 lieu of the smallest weights the T ^th and the yo^^h- 
 These riders are thus employed : one weighing y^th of a 
 grain is placed on the cross-piece of the extremity of the 
 sliding rod just mentioned, and the rod thus furnished is 
 brought gradually along the beam from the centre to the 
 end, until the rider can be deposited on the division on 
 the beam marked 10 ; the balance is then loaded on that 
 side with a weight equal to T V^h of a grain. If now the 
 rod be advanced to the centre of the balance, and the rider 
 dropped on the mark 5, the half of T Vth of a grain will be 
 pressing on that side of the balance, or, in other words,, 
 05 of a grain ; and when the rider is at the marks 1, 2, 
 3, 4, respectively, '01, -02, -03, -04 of a grain will be 
 indicated. With a rider weighing y^^th of a grain thou- 
 sandths of grains may be indicated : thus the last rider 
 placed on the marks 1, 2, 3, 4 would equal -01, -0002 y 
 003, -004 grain, &c. 
 
 THE METHOD OF WEIGHING. The operation of weighing is 
 very simple ; but as in the hands of the assayer it becomes 
 one of great frequency, the facilities for its performance 
 require to be mentioned. It should in the first place be as- 
 certained before every operation that the balance is in order, 
 so far as relates to its freedom of vibration, and also that 
 no currents of air are passing through the case, so as to 
 affect its state of motion or rest, a situation being chosen 
 where such influence may be avoided. In most cases there 
 is a small projecting arm on the upper part of the beam, 
 which, being turned either to the right or left hand side 
 of the beam as required, serves to establish perfect equili- 
 
WEIGHING 37 
 
 brium. Perfect equilibrium is, however, a matter of no 
 consequence if the assayer observes one or two simple 
 rules. He should never on any account weigh by the direct 
 method, that is, he should never obtain the weight of a 
 substance by putting it at once into one pan and then 
 counterpoising it by adding weights to the other pan. 
 This method is only to be relied on when the balance is of 
 rare perfection, and is used by no one but the assayer him- 
 self. The plan of weighing by difference should invariably 
 be adopted. By this means the weight of any body can 
 be readily ascertained, no matter whether the arms of the 
 balance are of unequal length or the pans out of equilibrium. 
 
 In the first place, it should be a rule that one pan, 
 preferably the left, be reserved for the substance to be 
 weighed, and the other pan be set apart for the weights. 
 
 Supposing the weight of a portion of mineral is re- 
 quired. First place a clean watch-glass, or platinum 
 <capsule, in the left pan, and carefully ascertain its weight. 
 Let us suppose it weighs 106*347 grains ; now put the 
 mineral in the watch-glass and ascertain the united weight 
 of the two. This we will imagine comes to 763*776. By 
 subtracting the weight of the glass or capsule from this 
 we find the true weight of the mineral, which is 763*776 
 -106*347 = 657*429. The substance to be weighed must 
 never be put direct into the pan. By weighing in this 
 manner by difference, the errors arising from inequality 
 in the equilibrium or length of arms are eliminated. 
 
 Nothing should ever be weighed until it is perfectly 
 cold. It is also unadvisable to weigh anything immediately 
 after it is taken from a cold place to a warmer one, as the 
 substance in such case will act as a hygroscopic body, 
 and, by condensing moisture, will appear heavier than it 
 really is. 
 
 Powders are conveniently weighed by filling a small 
 stoppered tube bottle with them, then weighing the whole, 
 and, after pouring out the requisite amount of its contents, 
 reweighing the bottle and powder. The difference gives 
 the weight of powder used. This is a very convenient 
 plan if several portions of the same substance are required 
 
.38 WEIGHING. 
 
 for different analyses. The tube will require reweighing 
 each time after the quantities required for each analysis 
 are shaken out into the receptacles. 
 
 A delicate balance is always furnished with means of 
 supporting the pans independent of the beam ; and the 
 beam itself is also supported when required by other bear- 
 ings than its knife-edges, and in such a manner as to admit 
 of the rapid removal of these extra supports when the 
 instrument is to be free for vibration. This is done that 
 the delicate edges of suspension may not be injured by 
 being constantly subjected to the weight of the beam and 
 the pans, and that they may suffer no sudden injury from 
 undue violence or force impressed upon any part of the 
 balance. When, therefore, a large weight of any kind is 
 put into or removed from the pans, it should never be 
 done without previously supporting them by these contriv- 
 ances ; for the weight, if dropped in, descends with a 
 force highly injurious to the supporting edges ; also, if a 
 large weight be taken out without first bringing the pans 
 to rest, it produces a similarly bad effect. 
 
 The weights should not be put into the pan at random, 
 It is a mistake to suppose that time is saved by such a 
 plan. The highest probable weight should be added first, 
 and then the set should be gone through systematically 
 down to the smallest weight, retaining or removing each 
 weight in order according as it is too little or too much. 
 The exact weight of a body will be found in this manner 
 in far less time than would be required were the weights 
 added by guesswork. 
 
 When a weight is put in which is assumed to be nearly 
 equal to the substance to be weighed, the balance should 
 be brought to a state of rest, and should then be liberated 
 gradually by turning the handle, so as to leave the pans 
 wholly supported by the beam. The whole being on its 
 true centres of suspension, it will be observed whether the 
 weight is sufficient or not ; and the rapidity of ascent or 
 descent of the pan containing it will enable a judgment to 
 be formed of the quantity still to be added or removed. 
 
 Great care should be observed in recording the weight 
 
WEIGHING. 39 
 
 in the notebook. The weight should first be ascertained 
 from an inspection of the vacancies in the box of weights, 
 and then verified by an examination of the weights them- 
 selves. This is conveniently done whilst replacing them 
 in the box, which should be done immediately after each 
 weighing. 
 
 In some cases, where great accuracy is not of so much 
 importance as rapidity in getting out approximate results, 
 a plan may be adopted recommended by Mr. F. F. Mayer 
 in the American c Journal of Science and Art ' for .1861. 
 
 He washes the precipitate thoroughly by decantation, 
 and then introduces it carefully into a bottle the exact 
 weight of which when filled with distilled water at a cer- 
 tain temperature is known. Since the precipitate is heavier 
 than water, the bottle, when filled again, will weigh more 
 than without the precipitate, and the difference between 
 the two weights furnishes the means of calculating the 
 weight of the precipitate. 
 
 In case the precipitate settles but slowly, it may be 
 collected on a filter, and, together wdth a filter, after wash- 
 ing, be introduced into the bottle, in which case the weight 
 of the filter and its specific gravity, supposing any differ- 
 ence should exist between its own and that of water, is to 
 be taken into account. Precipitates soluble in or affected 
 by water may be weighed in some other liquid. 
 
 Mr. Mayer applied this principle on a large scale as far 
 back as 1855. 
 
 In that year he was engaged in the manufacture of 
 lead carbonate from refuse lead sulphate, by treating the 
 latter, in a pulpy condition, with sodium carbonate. The 
 lead sulphate used contained very varying proportions of 
 water and soluble impurities, from which latter it had 
 first to be freed by washing. It was then in the state of a 
 thin pulp, and the difficulty was to find the amount of the 
 dry lead sulphate, as it was a matter of importance to use 
 as little sodium carbonate, and to obtain as pure a lead car- 
 bonate and sodium sulphate, as possible. This could only 
 be done by weighing it in the bulk or in portions ; but as 
 the drying of a tubful of lead sulphate (from 500 to 1,200 
 
40 WEIGHING MOIST PRECIPITATES. 
 
 Ibs.) was impracticable, and sampling not less so, since 
 the upper strata contained a much larger proportion of 
 water than the lead at the bottom, the following method 
 was contrived, which enabled the management of the pro- 
 cess to be left in the hands of a workman : 
 
 A strong oaken pail was taken, weighing 8 Ibs. when 
 empty, and a black mark was burnt in horizontally around 
 the inside of the pail two inches below the rim, up to 
 which mark it held 20 Ibs. of water. The specific gravity 
 of lead sulphate being 6-3, the pail, if filled up to the 
 mark, would hold 126 Ibs. of pure lead sulphate. The 
 specific gravity of water being 5*3 less than that of lead 
 sulphate, it followed that if there were 1 Ib. of water in the 
 pailful of moist sulphate, the pail would weigh 5- 3 Ibs. less 
 than 126 ( + 8, the tare of the pail) =120-7 ( + 8) ; if there 
 were 2 Ibs. of water present, the weight would be 115-4 
 ( + 8), and so on. This enabled a table to be calculated 
 giving in one column the actual weight of the pail when 
 filled with moist sulphate, and opposite, in a second column, 
 the amount of dry sulphate corresponding to the gross 
 weight. The weight of dry sulphate was thus found as 
 accurately as could be desired, although the amounts 
 varied in practice from 30 to 105 Ibs. 
 
 This is nothing but an application of the Archimedean 
 theorem, that when a solid body is immersed in a liquid 
 it loses a portion of its weight equal to the weight of the 
 fluid which it displaces, or to the weight of its own bulk of 
 the liquid. 
 
 Hence the rule, which is of great convenience in 
 volumetric analysis, that to find the weight of a moist 
 precipitate which is a compound of known specific gravity, 
 weigh it in a specific gravity bottle or some other vessel 
 of known weight when filled with water or any other 
 liquid at the normal temperature ; again fill it with the 
 water or other liquid, divide the excess of the new weight 
 by the specific gravity of the substance, less that of the 
 water or other liquid (that of water being =1), and add 
 the quotient to the overweight, which gives the weight of 
 the precipitate. 
 
INCINERATING PRECIPITATES. 41 
 
 INCINERATING PRECIPITATES PREVIOUS TO WEIGHING THEM. 
 A previous complete drying of the precipitate, as Bunsen 
 has proved, is in most cases not merely a loss of time but 
 a disadvantage ; whilst introducing the still moist precipi- 
 tate into the crucible requires the application of a very 
 gentle heat at the outset, and thus insures the most 
 favourable conditions for the easy and complete incinera- 
 tion of the filter-paper. Precipitates not washed upon the 
 filter-pump can be readily brought to a sufficient degree 
 of dryness if laid for a short time upon blotting-paper or 
 unglazed earthenware. 
 
 Dry filters may be also much better incinerated after 
 previous charring at the lowest possible temperature than 
 by rapid carbonisation or direct ignition in the flame. 
 How advantageous it is to char previously very slowly 
 may best be seen on incinerating filters whose contents 
 impede the complete combustion of the paper by the old 
 process, e.g. silicic acid, ammonio-magnesium phosphate, 
 &c. Charred paper obtained by rapid heating is deep 
 black and of a silky lustre, whilst if slowly carbonised it 
 is brownish-black, dull, and smoulders away like tinder. 
 Charred paper of the first kind appears under the micro- 
 scope perfectly amorphous, whilst the other displays the 
 carbonaceous skeleton of the fibre. 
 
 A careful removal of precipitates from the filter with 
 the exception of cases like zinc and cadmium, where 
 volatile reduction-products may be formed is quite use- 
 less, since the errors which it was hoped to obviate are 
 not really avoided. On incineration with the filter, wet 
 or dry, an error due to reduction may be easily corrected, 
 e.g. in barium sulphate with sulphuric acid ; in lead sul- 
 phate with nitric and sulphuric acid ; in iron and copper 
 oxides with nitric acid ; in silver chloride with nitric and 
 hydrochloric acids, &c. 
 
CALCINATION. 
 
 CHAPTEE III. 
 
 GENERAL PEEPAEATORY CHEMICAL OPERATIONS. 
 
 CALCINATION. Strictly speaking the term calcination means 
 the production of an oxide or Calx by combustion, and 
 it necessarily involves the intervention of atmospheric 
 oxygen. But in a metallurgical sense the term is re- 
 stricted to the separation of any volatile matter from a 
 mineral substance by the aid of heat alone, the atmosphere 
 being totally or partially excluded ; or the production of 
 rapid changes of temperature, so as, for instance, to render 
 minerals more fragile by heating and then quenching in 
 water, &c. 
 
 Thus we speak of the calcination of minerals, as iron 
 or zinc ores, &c., whose matrices are argillaceous, to 
 expel water, and also of gypsum to expel water ; calcium, 
 iron, copper, and lead carbonates are calcined to separate 
 carbonic acid ; zinc and iron hydro-carbonates, to get 
 rid of both water and carbonic acid ; cobalt and nickel 
 ores, &c., to separate arsenic and sulphur. The iron ores 
 found in the vicinity of collieries are calcined to expel 
 bituminous matter, and wood and bones to expel volatile 
 organic matter. Where the operation is accompanied by 
 combustion, and requires the oxygen of the atmosphere, 
 it is termed roasting. 
 
 Crucibles are conveniently used in calcination, as no 
 stirring of the mass is required. They may be made of 
 various materials, as clay, plumbago, platinum, silver, and 
 iron. Silver must not be employed when sulphur is pre- 
 sent, and it must not be exposed to a heat greater than 
 dull redness. The selection of the crucibles must depend 
 upon the substance under operation ; they must all be 
 furnished with covers. 
 
CALCINATION. 43 
 
 In almost all operations in assaying it is necessary to 
 estimate the amount of volatile matter lost by calcina- 
 tion. A very high temperature is seldom required in 
 calcination ; usually an air-furnace will give enough heat. 
 When the operation is finished the crucible must be 
 removed from the fire and allowed to cool gradually. 
 When completely cold, remove the cover and take out 
 the contents by means of a spatula. If any adhere, a 
 small brush will be found very useful for its removal. 
 The difference in weight before and after calcination will 
 represent the volatile matter. 
 
 When the subject to be calcined is fusible, the crucible 
 and contents must be weighed before ignition ; the loss of 
 weight is equal to the quantity of volatile matter expelled ; 
 in fact, this latter is usually the most satisfactory method 
 of conducting the experiment. 
 
 If the ignited substance be soluble in water, it can be 
 removed from the crucible by that means, employing heat 
 if required ; if not, any suitable acid may be used. 
 
 If the substance to be calcined decrepitates on heating, 
 it must be previously pulverised, and heated slowly and 
 gradually in a well- covered crucible. 
 
 Certain substances, as lead carbonate, undergo a 
 material alteration by contact with the gases given off 
 during the combustion of the fuel in the heating furnace ; 
 others, such as carbonaceous matters, are consumed by 
 the introduction of atmospheric air. All such substances 
 must be calcined in a closely covered crucible placed in a 
 second crucible (also covered) for further protection. 
 
 In some rare cases, however, these precautions are not 
 sufficient. In such, either a weighed porcelain or German 
 glass retort must be employed. 
 
 Sometimes earthenware crucibles lined with charcoal 
 are employed in calcination ; for even if the substance be 
 fusible, it may generally be collected and weighed without 
 loss, as very few bodies either penetrate into or adhere 
 to a charcoal lining. In this way grey cobalt and other 
 arsenio-sulphides are calcined at a high temperature to ' 
 expel the greatest possible amount of arsenic and sulphur. 
 
44 ROASTING OEES. 
 
 The selection and proper management of crucibles will 
 be given in the next chapter. 
 
 BOASTING. In this operation carbon, sulphur, selenium, 
 antimony, and arsenic are separated from certain metals 
 with which they were combined. Eoasting differs from 
 calcination in this particular : the latter is carried on in 
 close vessels, independent of the atmosphere ; the former, 
 in open vessels by the aid of the atmosphere. It is thus 
 we are enabled to separate the bodies just mentioned by 
 this process ; for the oxygen of the air, by combining 
 with them, forms a volatile substance which the heat 
 expels. Thus, in roasting copper and iron sulphide 
 (copper pyrites), 'the sulphur, copper, and iron mutually 
 combine with oxygen' to form sulphurous anhydride (vola- 
 tile), copper protoxide, and iron peroxide, thus : 
 
 2(FeS + CuS) + 130 = Fe 2 3 + 2(CuO) + 4(SO S ). 
 
 This is the final change in this case. During the process, 
 however, some copper and iron sulphates and sub-sul- 
 phates are formed. This change will be explained under 
 the head of Copper Assay. 
 
 When carbonaceous matters are roasted, the operation 
 also takes the name combustion, or incineration ; because 
 the object of roasting a fuel, for instance, is generally to 
 ascertain the amount of ash left. 
 
 In roasting, in the ordinary acceptation of the term, 
 the body must not be fused, but kept in a pulverulent 
 state ; there are, however, some cases in which fusion is 
 allowable, as in cupellation and .scarification. 
 
 The process of roasting is performed in different ways. 
 In one, a small flat vessel, called a roasting-dish (fig. 14), 
 is employed, made of the same material as the earthen 
 crucibles, and similar to a saucer. It is most conveniently 
 heated in muffle. The substance to be roasted must be 
 finely pulverised, placed in the roasting-dish, and con- 
 stantly stirred with an iron or glass rod until no fumes 
 are given off, or until it ceases to evolve the odour of 
 sulphurous acid if sulphur is one of the constituents to be 
 eliminated. 
 
ROASTING ORES. 45 
 
 The operation may also be performed in a crucible, 
 in which case it must be inclined to the operator, so 
 that the draught of air passing to the furnace flue may 
 impinge as much as possible on the substance under 
 manipulation. 
 
 During roasting the heat must be carefully regulated 
 for some time. At first it ought only to be the dullest 
 red ; and the substance must be assiduously stirred in 
 order to present the largest possible surface to the action 
 of the atmosphere and prevent fusion, for some assays, 
 when roasting, will fuse readily at a low temperature 
 unless the surface be continually renewed. Even by 
 paying the utmost atten- FlG 14 
 
 tion to this point it can- 
 not be always prevented, 
 as, for instance, when 
 antimony sulphide is 
 being roasted. In these 
 cases the assay must 
 be mixed with its own 
 weight of powdered quartz or fine white sand (silver 
 sand) ; the operation will then proceed steadily. 
 
 If the assay at all agglutinates, it must be taken from 
 the fire and rejected if the substance be plentiful ; if not, 
 the fused mass must be carefully removed from the cru- 
 cible or dish, pulverised, and the roasting recommenced. 
 In this case, however, the operation is always very tedious, 
 and the final result less exact, so that great care ought to 
 be taken at the commencement of the roasting. 
 
 When the assay has been kept at a dull red heat for 
 some time, and shows no signs of agglutination, the heat 
 may be slightly increased ; at the same time stirring must 
 be diligently pursued. After the heat has arrived at full 
 redness there is little fear of fusion ; and as the operation 
 proceeds more rapidly at a high temperature than at a low 
 one, it is well now to increase the heat to a yellowish red, 
 and even in certain cases to nearly a white heat. If the 
 stirring of the assay has been constant during the various 
 gradations of heat, the roasting at this point will be 
 
46 ROASTING OEES. 
 
 accomplished ; and the remaining operations of the assay 
 may be proceeded with. 
 
 This is the general plan of operation, but different sub- 
 stances require for roasting a different degree of heat ; for 
 instance, copper pyrites require a higher temperature than 
 grey copper ore, and the heat employed must in every 
 case be adapted to the substance to be roasted. Some 
 substances, for instance, arseniates, lead sulphate, &c., 
 cannot be roasted by heat alone. These require the 
 addition of a carbonaceous body to remove the combined 
 oxygen and allow the arsenic, sulphur, &c., to be com- 
 pletely roasted off. Ammonium carbonate in some cases 
 is also added to the mixture to decompose the sulphates 
 formed during the roasting of sulphides. 
 
 In cases where the metallic bases of the sulphides are 
 volatile, either as such or as oxides, as for instance galena, 
 antimony sulphide, &c., a loss of metal will always result 
 during the roasting process. 
 
 It may be as well to mention here that platinum cap- 
 sules are useful in certain roasting operations. Copper, 
 iron, and molybdenum sulphides are conveniently oxidised 
 in this kind of vessel, without much fear of injury, pro- 
 vided fusion of the roasting substance be carefully avoided. 
 Platinum vessels should also be used in ascertaining the 
 amount of ash in coal. 
 
 REDUCTION.- The process of reduction consists in re- 
 moving oxygen or an analogous element from any body con- 
 taining it, usually by means of either carbonaceous matter, 
 hydrogen, or a body containing both these elements, and 
 leaving the metal behind, usually in the form of a melted 
 button. The rationale of the operation is as follows, when 
 lead oxide is reduced with carbon : 
 
 2(PbO) + C = 2Pb + C0 2 . 
 
 In this case we start with lead oxide and carbon, and as 
 a result we obtain metallic lead and carbonic acid. 
 
 The reaction between nickel oxide and hydrogen is 
 thus expressed : 
 
 MO + 2H = Ni + H 2 0. 
 
REDUCTION. 47 
 
 Here we have at the commencement nickel oxide and 
 hydrogen ; and after the conclusion of the operation there 
 remains metallic nickel, and water which has volatilised. If 
 the reducing substance contain both carbon and hydrogen 
 the action will be thus, when a metal (e.g. lead) is reduced 
 from its oxide, carbonic acid and water being formed : 
 
 3(PbO) + CH 2 - 3Pb + C0 2 + H 2 
 
 In the operation of reduction by the aid of carbona^ 
 ceous matters two methods are employed : in the one, 
 charcoal, coal, sugar, starch, or any carbonaceous or 
 hydro-carbonaceous body, as argol, is mixed with the 
 substance to be reduced ; in the other, the process of 
 cementation is employed. Where sulphides are to be 
 reduced, metallic lead or iron is usually employed to 
 remove the sulphur. Generally, however, the sulphides are 
 previously converted into oxides by the operation of 
 roasting, and the reduction is then effected by means of 
 carbonaceous matter. 
 
 The process of cementation is conducted by placing 
 the oxide to be reduced in a crucible lined with charcoal, 
 and covering it closely while it is in the furnace ; the re- 
 duction proceeds gradually from the outside of the oxide 
 to the centre of the mass. The time requisite for this opera- 
 tion depends on three circumstances viz. the nature of the 
 oxide, the degree of temperature, and the mass acted on. 
 
 Some oxides treated this way are reduced very readily ; 
 others, again, take a considerable time ; while certain 
 of them do not appear to be acted on beyond the outer- 
 most layer. Of the first class is nickel oxide ; of the 
 second, manganese oxide ; and of the third and last, chro- 
 mium oxide. 
 
 Each of these classes of reduction has its advantages. 
 The former, or reduction by mixture with carbonaceous 
 matter, takes place very quickly and completely, but the 
 reduced metal is often mixed with carbon ; in the latter 
 process the residue is comparatively pure, but it is not 
 ' generally preferred, on account of the time and high tem- 
 perature necessary. 
 
48 REDUCTION. 
 
 Eeduction by hydrogen gas is very seldom employed ; 
 it is, however, necessary in some cases, as for instance in 
 the estimation of the percentage of cobalt or nickel in 
 a sample, where perfect accuracy is desirable. The opera- 
 tion is carried on in a tube of hard German glass, having 
 a bulb blown in its centre, which is heated either by a 
 spirit or gas lamp. Attached to it is a tube full of dried 
 calcium chloride, through which the hydrogen gas effect- 
 ing the reduction passes to perfectly dry it. 
 
 The bulb tube is weighed and the oxide introduced into 
 it ; it is again weighed, and the apparatus united by caout- 
 chouc tubes ; hydrogen gas (see Eeducing Agents) is then 
 passed through it until the whole of the atmospheric air 
 is expelled. Heat is afterwards applied till the bulb is 
 bright red, and the current of gas continued until no more 
 water from the decomposition of the oxide is formed ; the 
 source of heat is then removed, and the current of gas 
 continued until the apparatus is cold. The bulb tube, 
 with the reduced metal, is then weighed, and the amount 
 which it has lost represents the oxygen which the hy- 
 drogen has removed. By subtracting this oxygen from 
 the original weight of the substance, the difference gives 
 the amount of metal in the amount of oxide operated on. 
 
 FUSION. This operation is sufficiently simple, and is 
 employed in all assays by the dry way, in order to obtain, 
 in conjunction with the last process, a button or prill, as 
 it is termed, of the metal whose assay is in progress. It is 
 also a necessary step in the granulation of metals, the 
 preparation of certain fluxes and alloys, also lead for the 
 assay for silver, in order that a homogeneous ingot may 
 be obtained. Some ores, such as those of copper, are 
 melted instead of being roasted or calcined, in order to 
 prepare them for reduction. Minerals are also melted per 
 se, or with the addition of borax or sodium carbonate, in 
 order to ascertain the best treatment to be adopted in a 
 subsequent operation. Metals too are frequently melted 
 to drive off other volatile metals ; in this case the heat 
 should be continued for some time, and should be very 
 high, as it is difficult to remove the last traces of volatile 
 
SOLUTION. 49 
 
 metals. Thus, in melting the spongy gold left behind in 
 the retort after the distillation of gold amalgam, the ingot 
 of gold almost always retains mercury, which can only be 
 removed by repeated meltings at a very high temperature. 
 In some cases the fusion is intended to be only partial, 
 the object being to melt out an easily fusible part of the 
 mineral for instance, in assaying grey antimony ore and 
 different bismuth ores. 
 
 SOLUTION. In all cases where analysis in the wet way 
 is required, the mineral must be either wholly or partially 
 brought into the state of solution. The choice of a solvent 
 necessarily depends upon the nature of the material under 
 treatment. In some few cases water will be sufficient ; but 
 in the majority acids are required. Sometimes advantage 
 will be derived by first extracting all that water will dis- 
 solve, and then applying acids to the residue. In speaking 
 of the minerals, &c., which require solution for their assay, 
 the most appropriate solvents will be pointed out. In all 
 cases heat promotes solution. 
 
 Solution is best effected in glass flasks ; clean Florence 
 oil flasks are very appropriate for most purposes. They 
 may be supported on a hot sand bath, or on a metal ring 
 or coarse wire gauze, over the naked gas or spirit-flame. 
 The flask should then be placed in a sloping position, so 
 that when the liquid boils or effervesces from the escape of 
 gas, the drops spirted up may strike against the sloping 
 side, and run back into the liquid instead of being thrown 
 out of the mouth. 
 
 A porcelain dish may also be used, although from the 
 great surface exposed these vessels are more appropriate 
 for evaporation than solution. Beakers may likewise be 
 employed, but they should be covered over with an in- 
 verted funnel sufficiently large to rest within the top edge 
 without slipping down more than about half an inch ; or 
 a large watch-glass or dial-plate turned concave side up- 
 wards may be used as a cover. Both the funnel and dial- 
 plate serve the double object of keeping out dust and 
 preventing loss of the liquid by projection of fine drops 
 during ebullition. 
 
 E 
 
50 
 
 GLASS AND PLATINUM FORCEPS. 
 
 In many cases solution of the whole or part of a 
 mineral must be preceded by its fusion at a high tempera- 
 ture with sodium carbonate, nitre, or some other flux. 
 The fused mass must then be well extracted by boiling 
 water, when the residue will usually be found soluble in 
 hydrochloric or other acid. Special instructions in this 
 FlG 15 F IG> 16> method of effecting solu- 
 
 tions will be given in those 
 cases where it is necessary. 
 Where it is necessary 
 to manipulate in acid or 
 other solutions, the glass 
 and platinum forceps de- 
 scribed by David Forbes, 
 F.E.S., in the 'Chemical 
 News' for October 2, 1868, 
 will be found useful. 
 
 The accompanying 
 woodcut, fig. 15, shows 
 them in front and side 
 view, and will require but 
 little explanation. They 
 are made as follows : an 
 ordinary pair of strong 
 surgical forceps are taken 
 and the points cut off; a 
 small piece of sheet brass, 
 bent into a cylinder, is 
 then soldered to each arm 
 as shown at a ; these cylin- 
 ders being formed by merely bending the brass round, so 
 as to leave an open slit about one twentieth of an inch 
 wide in front. Two glass rods, such as are used for 
 stirrers, as long as the glass arms of the forceps are in- 
 tended to be, and just as thick as will enter these brass 
 cylinders when pressed with some force, are rounded by 
 the blow-pipe at the one end, whilst the other, when 
 softened, is somewhat flattened between the glassblower's 
 pliers, as seen in the woodcut. In order to complete the 
 
DISTILLATION. 51 
 
 forceps it is now only necessary to push each of these rods 
 into its corresponding brass cylinder or socket, the longitu- 
 dinal slits of which, by imparting a certain amount of 
 elasticity to the sockets, cause them to grasp the glass rods 
 firmly, and retain them without any cement or other fixing. 
 The relative lengths of the arms are easily adjusted by 
 slipping one rod more or less forward, whilst the points 
 can be made to hold and meet accurately, by rubbing 
 them down on a piece of sandstone. 
 
 Such forceps may, of course, be made to any convenient 
 size ; the one figured in the woodcut is drawn to exactly 
 half-size, and is found to be of very useful dimensions for 
 general analytical work, especially when manipulating in 
 nitro-hydrochloric acid, nitrate of silver, and other solutions 
 which would have acted upon metals, horn, ivory, &c. 
 
 Fig. 16 represents another convenient form of forceps, 
 also drawn to one half the real size, with long platinum 
 points soldered to the steel body at a\ these have been 
 also found of great service in general laboratory operations, 
 especially when hydrofluoric acid is in question. 
 
 DISTILLATION. There are two distinct classes of this 
 operation : in the one, liquids are submitted to experiment 
 with the object generally of separating them from sub- 
 stances which are non-volatile, and will consequently be 
 left behind when the liquid comes over. Belonging to 
 this class may be mentioned the distillation of nitric acid, 
 the preparation of distilled water, and the separation of 
 mercury from gold and silver amalgam. In the other 
 kind of distillation, which goes by the name of dry distil- 
 lation^ solid bodies, as wood, coal, &c., are subjected to 
 heat in order generally to ascertain the amount of gas or 
 other volatile matter given off in the course of an experi- 
 ment, from a certain quantity of the coal or other sub- 
 stance operated upon. 
 
 In liquid distillation (as in the purification of nitric 
 acid, &c.), retorts are used. The best form for general use 
 is that which is furnished with a stopper at the upper 
 part of the body, a (fig. 17), through which the liquid is 
 introduced ; the neck of the retort is then placed in that 
 
 E 2 
 
52 
 
 DISTILLATION 
 
 FIG. 17. 
 
 of a receiver, b, over which a piece of wet cotton or 
 woollen cloth is placed, and which must be kept cold 
 
 by means of a stream of water 
 from a funnel, c, the shaft of 
 which is partially plugged up 
 with cotton wool. Heat is then 
 applied to the retort, and as 
 much of the liquid as is desired 
 is distilled over into the re- 
 ceiver. It is advisable not to 
 fill the retort more than two 
 thirds full, and to apply the 
 heat at first very gently, other- 
 wise there is a risk of breaking 
 the vessel. 
 
 A more convenient form of apparatus for distillation 
 and condensation is shown at fig. 18, in which a Liebig's 
 
 FIG. 18. 
 
 condenser is attached to the retort. Fig. 19 will show 
 the construction of the condensing apparatus. The cold 
 water passes into the funnel above, is conveyed at once 
 to the lowest end of the condenser, whilst the heated water 
 passes off by the upper tube. 
 
 Distilled water is a most important agent in the labora- 
 tory ; and, as much is needed, it is better to have a still 
 
DISTILLATION. 
 
 53 
 
 specially adapted for its production. Such a one is de- 
 picted at fig. 20, where A is the body of the still ; B the 
 
 furnace in which it is 
 
 , /,i ,.-,-. ! , FIG. 19. 
 
 set (the still may also be 
 
 placed in the portable 
 
 furnace, fig. 25, p. 65) ; 
 
 C the still head ; D E 
 
 the neck ; F the worm ; 
 
 I J K L the worm-tub 
 
 containing cold water to condense the steam generated in 
 
 the still ; M N the pipe to lead fresh cold water to the 
 
 FIG. 20. 
 
 bottom of the worm-tub, while the warm water runs off 
 .at the top, as in Liebig's condenser ; and P the vessel in 
 which the distilled water is received. 
 
 In the dry distillation of bodies, earthenware, glass, or 
 iron retorts are employed ; but for small operations a tube 
 of wrought-iron, about one inch internal diameter, and 
 plugged at one end, is found to be a convenient form of 
 apparatus. It is placed with the substance contained in it 
 in a furnace, and a small tube, either of glass or pewter, 
 is fixed by means of a perforated cork to the open end of 
 the large tube. The gas given off during the operation 
 may be collected by the aid of a pneumatic trough. 
 
64 SUBLIMATION. 
 
 SUBLIMATION. This operation is a kind of distillation- 
 in which the product is obtained in the solid form. The 
 apparatus which may be employed for this purpose are 
 tubes, flasks, capsules, or crucibles. Florence flasks are 
 exceedingly useful ; they may be sunk in a sand bath, and 
 the sublimed substance received directly into another flask, 
 or by passing through an intermediate tube. Sometimes, 
 however, it is difficult to entirely remove the sublimed sub- 
 stance ; and in order to avoid this inconvenience, Dr. Ure 
 has proposed the following very excellent subliming appa- 
 ratus : It consists of two metallic or other vessels, one of 
 which is flatter and larger than the other. The substance 
 to be sublimed is placed in the smaller vessel, and its 
 opening is covered by the larger filled with cold water, 
 which may be replaced from time to time as it becomes 
 hot. The sublimed substance is formed on the lower part 
 of the upper vessel. A large platinum crucible, filled with 
 cold water, and placed on the top of a smaller one, answers 
 the purpose of the before-mentioned apparatus very well. 
 
 SCORIFICATION CuFELLATiON. These operations will be 
 described under the head of Silver Assay. 
 
FURNACES. 55 
 
 CHAPTEE IV. 
 
 PRODUCTION AND APPLICATION OF HEAT. 
 
 FURNACES for assay purposes may be heated either by solid 
 fuel, oil, or gas, and they may be divided into wind and 
 blast furnaces. In the former the fire is urged by the or- 
 dinary draught of a chimney, and in the latter by means 
 of bellows or artificial blast. We shall commence with the 
 former, as they are in most common use. They are of 
 various kinds, according to the purposes for which they are 
 required. The three principal kinds are those for fusion, 
 calcination, and cupellation. Coal, coke, and charcoal 
 are the fuels employed, and the merits of each will be par- 
 ticularly discussed. Blast furnaces are only employed for 
 the purpose of fusion, although their forms are various ; 
 charcoal and coke are the fuels most in use, but oil and gas 
 blast furnaces are used in small laboratory operations, and 
 for many purposes they are preferable to other furnaces, 
 on account of their freedom from dust and dirt, and the 
 perfect control the operator possesses over the heat. 
 
 Furnaces consist of certain essential parts viz. first, 
 the ash-pit, or part destined to contain the refuse of the 
 combustible employed ; secondly, the bars on which the 
 fuel rests ; these are sometimes made movable, or are fixed 
 to a frame ; the former arrangement is more convenient, 
 as it allows clinker and other refuse matters to be readily 
 removed ; thirdly, the body of the furnace in which the 
 heat is produced ; and lastly, in wind furnaces, the chimney 
 by which the heated air and gaseous products of combus- 
 tion are carried off. 
 
 CALCINING FUKNACE. Calcining furnaces are small and 
 shallow, because a high temperature is not required. They 
 
56 CALCINING FURNACES. 
 
 may be made square or circular ; the former are most 
 readily constructed, and, where many crucibles are to be 
 heated at once, they are preferable to the circular ; but the 
 latter give the greatest degree of heat with the least pos- 
 sible consumption of fuel, and are to be preferred on that 
 account where one crucible only is to be ignited. 
 
 The body of the furnace is best made with good bricks, 
 lined with Welsh lump, fire-bricks, or a mixture of Stour- 
 bridge clay and sand. It is also desirable that a plate of 
 iron with a ledge be placed over the upper part of the 
 furnace to protect the brickwork from blows with crucible 
 tongs, &c., and to keep it in its place when disturbed by 
 sudden alterations of temperature. The bars of the fur- 
 nace may be either in one single piece, or made up of 
 several bars of iron fastened to a frame. They ought to 
 be as far as practicable from each other, and must not be 
 too large, although large enough not to bend under the 
 weight of the fuel and crucibles when they become hot, 
 and they must not be so far removed from each other 
 as to allow the coke or charcoal to fall through easily. 
 Lastly, the more readily the air can find access to the 
 centre of the fuel, the higher will be the temperature pro- 
 duced in the furnace ; very simple assays occasionally fail, 
 only because the bars are either too large or too close 
 together. 
 
 CHIMNEY. Calcining furnaces generally have no fixed 
 chimney, but are covered with a movable one when a 
 greater degree of heat is required. This chimney may be 
 about five feet high, the diameter of the furnace at the 
 bottom, and tapering off to about two thirds of that dia- 
 meter at the top. It is made of strong plate iron, fur- 
 nished with a wooden handle. The lower part is provided 
 with a door, by means of which the interior of the furnace 
 may be examined without disturbing the whole arrange- 
 ment of the chimney, and consequent cooling of the con- 
 tents of the furnace. 
 
 If, during the course of any experiment, noxious or 
 offensive vapours are expected to be given off, the furnace 
 must be so arranged that they may be introduced into a 
 
WIND FUKflACE. 57 
 
 flue, by fastening a piece of iron plate pipe, furnished with 
 an elbow joint, on to the movable chimney before spoken 
 of. 
 
 EVAPORATING FURNACES. The furnaces just described 
 answer exceedingly well in the absence of gas, for heating 
 small flasks, evaporating basins, &c., when surmounted by 
 a tripod stand or sand bath. This is necessary, as many 
 assays by the dry way are preceded and followed by cer- 
 tain operations in the wet way. 
 
 THE HOOD. In order to prevent certain gases or va- 
 pours from fires, evaporating basins, &c., from entering 
 into the laboratory, a large metal covering, termed a hood, 
 is employed, terminating in a chimney having a good 
 draught. It is best made of sheet or galvanised iron. 
 
 FUSION FURNACE WIND FURNACE. The wind furnace, 
 properly so called, is a furnace provided with a chimney, 
 and capable of producing a very high temperature. 
 
 Wind furnaces are generally square, but, if more than 
 four crucibles are to be heated at one time, they may be 
 made rectangular, the chimney being placed at one of the 
 long sides. When the furnace is required to hold but one 
 pot, it may, however, be made circular. 
 
 The body of the furnace ought to be made of good 
 bricks, solidly cemented with clay, and bound by strong 
 iron bands. The bricks must be very refractory, and 
 capable of sustaining changes of temperature without 
 cracking. They are ordinarily made with the clay used 
 in the manufacture of crucibles. In some cases bricks are 
 not used for the lining of this kind of furnace ; for in- 
 stance, a mould of wood is placed in the centre, and the 
 open space between the surface of that and the outer brick- 
 work is filled with a paste of very refractory clay, each 
 layer being well beaten down. When the space is filled the 
 case is withdrawn, and the crust of clay dried with much 
 precaution, every crack that may be caused by unequal 
 desiccation being filled up as fast as formed. This method 
 of manufacture is very applicable to circular furnaces. In 
 every case, however, it is necessary to border the edge 
 with a band of iron to prevent injuries from tongs, or pots. 
 
58 WIND FURNACE. 
 
 By using a mixture of 1 part of refractory clay and 3 to 
 4 parts of sifted quartz sand, no cracks are formed during 
 desiccation. This mixture is used on the Continent for the 
 interior fittings of Sefstrom's blast furnace, as well as for 
 larger blast furnaces for manufacturing purposes. It is 
 said to stand a high temperature exceedingly well. 
 
 Makins * recommends for small furnaces the second 
 kind of bricks, known as Windsor, or in the trade P.P. 
 bricks. ' These are of a red colour, very siliceous, but soft, 
 easily cut and shaped, and yet standing heat very well 
 The best method of cutting them is by a piece of zinc 
 roughly notched out as a saw, and then the more accurate 
 figure required may readily be given them by grinding 
 upon a rough flat stone. In this way the small circular 
 furnace formerly made by Newman, and sold by him as 
 his " universal furnace," is lined by cutting the bricks with 
 care to the radii of the circle they are to form, when they 
 key in, like an arch, and so need no lining whatever.' 
 
 THE ASH-PIT is an open space under the bars, which 
 serves as a receptacle for ashes, clinkers, &c., produced 
 during the time the furnace is in use. It should have the 
 same area as the furnace, and be completely open in front, 
 so that the air may have free access ; it is well, however, 
 for the sake of economy, to furnish this opening with a 
 hinged door, having a register plate fixed in it, so that the 
 draught may be reduced, or entirely shut off, in order that 
 the fire may be extinguished when desirable, and fuel 
 saved which would otherwise be burnt in waste. 
 
 On the one hand, it is well to have the power of cut- 
 ting off access of air into the body of the furnace by the 
 lower part, either to put out the fire entirely, or to dead en 
 it whilst putting in a pot ; and, on the other, to attain the 
 maximum of temperature, we must have the means of 
 allowing the air to pass with the greatest possible facility 
 into the furnace. In order to do this it is necessary to 
 furnish the ash-pit with doors, or valves, whereby the 
 quantity admitted may be regulated as desired. It is ad- 
 vantageous to lead the air to the ash-pit from a deep and 
 
 * Makiris's Metallurgy, p. 88. 
 
WIND FURNACE. 50 
 
 cold place, by means of a wide pipe. A chimney of less 
 height will then be required. 
 
 THE BARS are made in one piece, or are made up of 
 movable pieces of metal; the latter arrangement is the 
 most convenient. Wherever a wind furnace is in use, the 
 upper opening is closed by a cover made of a fire-tile r 
 encircled with iron. 
 
 THE CHIMNEY is a very essential part of a wind furnace : 
 it is on its height and size that the draught depends, and, 
 in consequence, the degree of heat produced within the 
 furnace. In general, the higher and larger the chimney r 
 
 FIG. 21. 
 
 the stronger is the draught ; so that, by giving it a great 
 elevation, exceedingly high temperatures may be obtained. 
 But there is a limit which it is useless to pass in a furnace 
 destined for operations by the dry way ; and, besides this, 
 the building a very high chimney presents many difficulties 
 and much expense, so that in laboratory operations, where 
 a very strong current of air is required, recourse is had to 
 a pair of double bellows. A temperature can be produced 
 in a wind furnace sufficiently strong to soften the most 
 
60 WIND FURNACE. 
 
 refractory crucibles, by means of a chimney from 
 six to forty feet high. 
 
 Chimneys are generally made square or rectangular, 
 and have interiorly the same dimensions as the body of 
 the furnace. About two feet above the upper part of the 
 furnace they are furnished with a register or damper, by 
 means of which the current, of air may be regulated or 
 entirely stopped at will. The damper is a plate of iron 
 sliding into a small opening across the chimney. 
 
 A wind furnace of the kind above described is repre- 
 sented by fig. 21. 
 
 The left-hand figure in 21 is the plan, the middle the 
 elevation, and the right is a sectional view. A the body 
 of the furnace in which the crucibles to be heated are 
 placed, G the bars, and P the ash-pit ; the cover is formed 
 of a thick fire-tile of the requisite size, firmly encircled by 
 a, stout iron band, and furnished with a handle for con- 
 venience in moving it ; B the flue, C the chimney, E the 
 damper ; H a hood over the furnace, supported by iron 
 bands h h h ; M the handle of a ventilator T 7 , which serves 
 to carry off hot air and fumes from furnace when open ; 
 and finally, /S, a small sand bath, in which to set the red- 
 hot crucibles when taken from the fire ; one foot square 
 inside the fireplace of the furnace is a good and convenient 
 size ; the remainder will then be in proportion. 
 
 BLAST FURNACES. In this species of furnace the air 
 necessary to keep up the combustion is forced through the 
 fuel by means of a blowing apparatus, instead of being- 
 introduced by the draught of a chimney as in the wind 
 furnace. 
 
 The most convenient apparatus for forcing air into a 
 furnace is a double bellows ; a fan may be used, but it is 
 not so powerful. 
 
 The quantity of air passing into a furnace varies with 
 the length of the assay, and ought to increase gradually 
 as the temperature becomes higher. 
 
 The following is the description of a most excellent 
 blast furnace which has been in use for some years in the 
 laboratory of the Royal Institution : The temperature 
 
BLAST FURNACE. 61 
 
 produced by it is extraordinary, considering the small 
 amount of time and fuel employed. It is sufficiently 
 powerful to melt pure iron in a crucible in ten or fifteen 
 minutes, the fire having been previously lighted. It will 
 effect the fusion of rhodium, and even pieces of pure 
 platinum have sunk together into one button in a crucible 
 subjected to its heat.* All kinds of crucibles, including 
 the Cornish and Hessian, soften, fuse, and become frothy 
 in it ; and it is the want of vessels which has hitherto put 
 a limit to its application. The exterior (fig. 22) consists 
 of a black-lead pot, eighteen inches in height, and thir- 
 teen inches in external diameter at the top ; a small blue- 
 pot of seven and a half inches external diameter at the top 
 has the lower part cut off so as to leave an aperture of 
 five inches. This, when put into the larger part, rests 
 upon its lower external edge, the tops of the two being 
 level. The interval between them, which gradually in- 
 creases from the lower to the upper part, is filled with 
 pulverised glass-blowers' pots, to which p IG< 2 2. 
 
 enough water has been added to moisten 
 the powder, which is pressed down by 
 sticks so as to make the whole a com- 
 pact mass. A round grate is then 
 dropped into the furnace, of such a size 
 that it rests about an inch above the 
 lower edge of the inner pot ; the space 
 beneath it, therefore, constitutes the air- 
 chamber, and the part above, the body 
 of the furnace. The former is 7-J inches from the grate 
 to the bottom, and the latter 7-^ inches from the grate to 
 the top. Finally, a longitudinal hole, conical in form, and 
 1^ inch in diameter in the exterior, is cut through the 
 outer pot, forming an opening in the air-chamber at the 
 lower part, its use being to receive the nozzle of the 
 bellows by which the draught is thrown in. 
 
 Sefstrom's blast furnace, obtainable at most chemical- 
 instrument makers, is also very powerful and convenient ; 
 it consists of a double furnace. It is made of stout sheet- 
 
 * Faraday. 
 
02 CUPEL FURNACE. 
 
 iron, lined with fire-clay, and is used with coke, or char- 
 coal and coke, broken into pieces of about a cubic inch in 
 size. The blast of air is supplied by a powerful blowing- 
 machine. It will readily produce a white heat. Indeed 
 the limit to its power seems to be the difficulty of finding 
 crucibles or interior furnace fittings which will stand the 
 temperatures produced in it without softening. Kersten 
 states that he increases the heat in Sefstrom's blast furnace 
 by using a hot blast. 
 
 H. Ste.-Claire Deville has employed for melting platinum 
 a furnace 12 inches high, and 11 inches wide, which 
 rests upon a cast-iron plate full of holes. This is connected 
 with a forge bellows. After blowing for a few minutes, 
 the temperature of the furnace will have reached the 
 highest possible degree, but this zone of maximum heat 
 only extends to a small height above the bottom of the 
 furnace. Above this point a considerable quantity of car- 
 bonic oxide gas is formed, which burns with a very long 
 flame. The heat produced in this furnace is so high that 
 the best crucibles melt, and only crucibles made of good 
 and well-burned lime can be used. 
 
 THE MUFFLE OR CUPEL FURNACE is a furnace in the centre 
 of which is placed a small semi-cylindrical oven, which is 
 termed the muffle. These furnaces were in use as early 
 as the thirteenth century. Their construction and dimen- 
 sions depend 
 
 1. On the temperature which the furnace is intended 
 to produce ; 
 
 2. On the number of cupellations required to be per- 
 formed at one time ; and 
 
 3. On the kind of fuel to be used. 
 
 The muffles are mostly made of refractory clay, gener- 
 ally of one piece, but it is best to make them of two pieces ; 
 the bottom being one, and the cover or vault the other. 
 
 Muffles are sometimes made of cast-iron, cast in one 
 piece. They are useful in small furnaces intended for 
 cupellations only. 
 
 Muffle furnaces must always be provided with a good 
 chimney to carry off the hot gases. 
 
CUPEL FURNACE. 
 
 
 The muffle, being completely surrounded by ignited 
 fuel, acquires a very high temperature, and in its interior 
 all operations requiring the presence of air, and which 
 cannot be carried on in contact with carbonaceous matters, 
 may be performed such as roastings, scorifications, and 
 cupellations. 
 
 When from ten to twenty cupellations ha.ve to be 
 effected at one time, large brick furnaces are employed ; 
 and, in consequence, much fuel is consumed to waste in 
 these when only a few cupellations are required. This 
 has occasioned many persons to endeavour to form small 
 furnaces, where one or two cupellations may be carried 
 on with the smallest possible quantity of fuel. 
 
 MM. Aufrye and d'Arcet have contrived a furnace 
 which is capable of fulfilling all these conditions. 
 
 The furnace is elliptical, about 7 inches wide and 18 
 high; its ash-pit has but one circular opening, and its 
 height is such that, when the furnace is placed upon it, 
 and .the whole upon a table, the assayer can, when seated, 
 readily observe the course of the assay within the muffle. 
 The hearth has five openings, in one of which the muffle 
 is placed ; in another a brick to support it ; a third is for 
 the purpose of introducing a poker to stir the ashes, and 
 make them fall through the grate-holes : this can be closed 
 with a small earthen plug ; and lastly, there are two round 
 holes, placed in its largest diameter, to facilitate the in- 
 troduction of air, either by draught or a pair of bellows, 
 as the case may require. The support for the fuel is 
 generally a plate of earthenware, pierced with holes, and 
 bound round with iron wire to keep it together in case it 
 cracks by changes of temperature ; but it is better to use 
 an iron grating. 
 
 The dome of the furnace has a circular opening, which 
 can be closed by a plug of earthenware ; this opening 
 serves for the introduction of the fuel. A chimney is 
 necessary to increase the draught; it is made of sheet-iron, 
 and may be from 1^ to 2 feet in height, and ought to fit 
 the upper part of the dome very exactly. At its base 
 there is a small gallery, also of sheet-iron, in which it is 
 
64 CUPEL FURNACE. 
 
 intended to place the new cupels, so that they may be 
 strongly heated before introduction to the muffle. This 
 saves many of them from fracture. 
 
 MM. Aufrye and d'Arcet have estimated the quantity 
 of charcoal necessary to heat this furnace. The following 
 are comparative experiments : 
 
 Silver Lead Time Standard Charcoal 
 
 employed, employed, of assay, used, 
 
 No. grains grains minutes thousandths grains 
 
 1 1 4 12 947 173 
 
 2 1 4 11 950 86 
 1 4 13 949 93 
 1 4 10 949 60 
 
 Coke or charcoal may be used in this furnace, but the 
 fire must be lighted by means of charcoal alone, as coke is 
 very difficult to inflame in a cold furnace. When it is red- 
 hot it may be fed with coke, or, better still, a mixture of 
 coke and charcoal. 
 
 Where great numbers of cupellations have to be made 
 at once, the following form of brick furnace is requisite : 
 
 Fig. 23 shows an elevation of the furnace ; fig. 24 
 shows a section. The interior of the furnace is of fire- 
 brick ; the exterior, of common brick. The upper part 
 is protected by a plate of iron, and the upper opening,, 
 through which the fuel is introduced, is covered, when 
 necessary, by a large fire-tile strongly encircled with an 
 iron band, to which are attached two handles, by which 
 the whole can be moved. 
 
 The muffle opening, as seen partially open in the dia- 
 gram, can be entirely closed by means of two sliding doors y 
 made of sheet iron, running in a stout wrought-iron frame, 
 built into the brickwork. Immediately below the muffle 
 entrance are two movable bricks ; these close the openings 
 through which the fire-bars are introduced ; and still 
 lower down is the ash-pit door, furnished with a register 
 for the better regulation of the current of air required by 
 the furnace. In fig. 24 is shown a brick built into the 
 back of the furnace, on which the close end of the muffle 
 is supported. This brick may, however, be replaced by a 
 crucible or fire-brick standing on the bars of the furnace: 
 
 A very useful furnace for small operations is one which 
 
UNIVERSAL FURNACE. 
 
 has been aptly termed the ; universal furnace,' as it is 
 capable of performing all that is required of any furnace 
 
 FIG. 23. 
 
 FIG. 24. 
 
 in an assay (see figs. 25 and 26,^elevation and section). Tt 
 is much to be recommended for its durability, cheapness, 
 
 FIG. 25. 
 
 FIG. 26. 
 
 and its small size compared with the heat it can produce. 
 It is constructed externally of sheet iron, very stout, and 
 
 F 
 
66 FURNACE OPERATIONS. 
 
 is lined with fire-brick, not cemented together, but ground 
 and keyed as an arch, so that it can never fall out till it 
 is completely useless. Its height is about 2J feet and 
 diameter 1 foot ; internal diameter 8 inches and depth of 
 fireplace 1J feet. It is furnished with five doors, one in 
 the ash-pit and four in the body of the furnace, two in the 
 front, one above the other, and two opposite each other, 
 at the sides. The cover serves as a sand-bath, and when 
 that is taken off there is a series of cast-iron rings, fitting 
 the top of the furnace, where basins can be placed either 
 for the purpose of evaporation, calcination, or roasting. 
 The two opposite holes serve for the introduction of a 
 tube in operations where it is requisite to pass a gas over 
 any body at a red heat. In the lower hole in front can be 
 placed a muffle for roastings and cupellations, introducing 
 fuel and crucibles by the upper one ; it also serves as an 
 opening through which the state of the furnace can be 
 seen, or the progress of any assay observed. 
 
 Iron, manganese, nickel, and cobalt can be fused in 
 this furnace when it has a flue of about thirty feet in 
 height attached to it, and by closing the ash-pit door the 
 dullest red heat, for gentle roastings, can be obtained. 
 
 FURNACE OPERATIONS. 
 
 Crucibles must be carefully supported in the fire, and 
 must always be covered. They must stand solidly, and 
 be at equal distances from the sides and bottom of the 
 furnace, so as to receive a like share of heat, and they 
 must be completely surrounded with fuel. If a crucible 
 is supported on the grate-bars of a furnace, the draught 
 of cold air will prevent the bottom from getting hot. If 
 it is supported on the fuel, the bottom gets heated quickly, 
 but the fuel on burning away allows the crucible to fall 
 down, and may cause the loss of the contents. For these 
 reasons a crucible should always be supported on a piece 
 of fire-brick about three or four inches high. In many 
 cases an old crucible inverted will serve as a convenient 
 support. The fire must be got up gradually, so as to 
 
FURNACE APPARATUS. 67 
 
 prevent the sides of the furnace and the crucibles within 
 from cracking from the sudden increase of heat. No time 
 is saved by urging the fire strongly at first, for crucibles 
 are bad conductors of heat, and a high temperature at 
 the commencement scarcely causes the heat to penetrate 
 to the interior faster than a moderate redness. After the 
 furnace has arrived at a full red heat, more air may be 
 given, and in from about twenty minutes to one hour 
 the assay will be finished. During the time that the fur- 
 nace is in full action, the cover must be occasionally re- 
 moved to add more fuel, if any open spaces occur round 
 the crucibles, also to press the fuel close to the pots. 
 When the pots are taken out they may be placed on the 
 anvil or in a sand-bath, and allowed to cool gradually 
 before they are broken to examine their contents. 
 
 In commencing a second assay immediately in the same 
 furnace, certain precautions must be taken to insure 
 success. In the first place, all ash and clinker must be 
 removed from the grate by means of a crooked poker ; 
 secondly, the fuel must be pressed down firmly ; and 
 lastly, a layer of fresh combustible must be placed on 
 the fire, and before that is ignited the crucibles must be 
 
 ' O 
 
 arranged upon the support and the spaces about them be 
 filled with coke or charcoal, as the case may be, and the 
 assay be proceeded with as before. 
 
 In executing many assays, one after the other, a great 
 saving of fuel is effected, for the furnace is not allowed 
 time to cool. 
 
 AUXILIARY APPARATUS. Ordinary assay furnaces require 
 very few instruments ; they are, firstly, pokers or stirring 
 rods, made of stout bar-iron ; these may be straight, as for 
 stirring the fuel from the top of the furnace, so as to fill 
 up cavities formed by uneven combustion; or curved, for 
 clearing the bars from below from clinker and ashes. 
 Straight and curved tongs are also required ; for small 
 crucibles the blacksmith's common forge tongs are the 
 most suitable ; tongs with semicircular ends (see fig. 27) 
 are very serviceable for larger crucibles. The tongs a are 
 particularly adapted for removing large cupels or calcining 
 
 F 2 
 
68 
 
 FURNACE TOXGS. 
 
 dishes from the muffle ; the tongs b and c are used for 
 lifting heavy crucibles from the wind or blast furnace. In 
 case the eyes of the operator are weak, it is advisable to 
 make use of a pair of deep neutral-tint spectacles. Most 
 of the radiant heat from the interior of a furnace may be 
 
 FIG. 27. 
 
 cut off by holding before the face a large sheet of window 
 glass'; or the operator may look at the reflected image in 
 a looking-glass, instead of looking direct into the furnace 
 itself. Some assayers recommend the use of masks for the 
 face and gloves for the hands, but these are not needed. 
 In cupel furnaces, both curved and straight pokers or 
 stirring rods are required ; also a curved rod made of 
 lighter iron, to be used in closing the sliding doors, slightly 
 moving cupels, &c. The tongs used vary in form (see 
 fig. 28). a represents very light elastic tongs or pincers 
 
 FIG. 28. 
 
 employed in the introduction of lead and other matters to 
 the cupel ; b shows the tongs for holding the scorifier ; 
 the curved part fits the lower part of the scorifier, and the 
 upper or single part passes over the upper part of the 
 scorifier, so that its contents may be emptied into the 
 proper mould without fear of its slipping from the opera- 
 
INGOT MOULDS. 
 
 tor's grasp ; c represents the tongs used in moving cupels ; 
 they are slightly curved, so that the cupels from the back 
 part of the muffle may be removed without disturbing those 
 
 FIG. 29. 
 
 in front. Pig. 29 shows the plan and section of the ingot 
 mould, into which the contents of the scorifiers are poured ; 
 it is made of thin sheet iron, and the depressions for the 
 reception of the fused lead slag^ and ore are hammered out. 
 
 FIG. 30. 
 
 Fig. 30 is a wrought-iron ladle, in which lead clippings, 
 &c., are melted, in order to obtain a fair average of a large 
 Fio 31 , quantity ; and fig. 31 represents 
 
 the ingot mould into which the 
 fused lead or other metal is 
 poured. Other special apparatus 
 will be described under the assay 
 in which they are required. 
 
 Furnaces are heated with 
 anthracite, coke, and charcoal, and sometimes with a 
 
70 FUEL FOR FURNACE. 
 
 mixture of the two latter ; coal is very seldom employed,, 
 and therefore will not be much spoken of; coke is the 
 principal combustible used in assaying. Calcining fur- 
 naces ought to be heated with charcoal alone, because 
 coke employed in small quantities ignites and burns with 
 too much difficulty. All fuel, contains certain fixed matters 
 which remain after combustion, and which constitute the 
 ash. This ash fuses or agglutinates together, and when a 
 certain quantity is formed, if it be not removed, the fire 
 will decrease in intensity, and finally die out. As all com- 
 bustibles do not contain the same amount of ash, they 
 should be carefully selected ; those containing the least 
 are to be preferred, in the first place because, weight for 
 weight, they contain more available fuel, and, secondly,, 
 because they can be used in a furnace a longer time with- 
 out the formation of so much clinker. The composition 
 of the ash likewise merits much attention. 
 
 Charcoal contains, in general, from 3 to 4 per cent, of 
 ash, the chief components of which are lime and potash 
 carbonates. Certain other matters are also present, as 
 phosphoric acid, iron oxide, manganese, &c., but these 
 are in very minute proportions. The ash is not fusible 
 per se, and, if it does not meet with any substance capable 
 of combining with it, it passes through the bars as a white 
 powder ; but when the potash predominates, it exercises 
 a corrosive action on the bricks with which the furnace is 
 lined, as also on crucibles, lutes, &c., by the formation of 
 a fusible potassium silicate, which in course of time runs 
 down the sides of the furnace, and chokes the bars. 
 Whenever the ash is in very small proportion to the 
 charcoal, its fusion is rather useful than otherwise, be- 
 cause it forms a species of varnish, which, penetrating the 
 surface of the bricks and lutes, gives them solidity by 
 binding them together with a cement, forming part of 
 their substance. 
 
 The proportion of ash which coke contains is very 
 variable; some commercial samples contain from 8 to 10 
 per cent., while others, made from very pure coal, give 
 but 2 to 3 per cent. ; so that this fuel also ought to be 
 
CHARCOAL COKE. 71 
 
 carefully chosen. The nature of this ash is different from 
 that of charcoal ; it consists principally of iron oxide and 
 clay. The former is produced from the pyrites which 
 coal generally contains. The clay is similar to the car- 
 bonaceous schists, not very fusible by itself, but neverthe- 
 less capable of softening, When pure, it forms a slag, 
 which attacks neither the bricks nor crucibles. This 
 happens very rarely ; it is more often that iron oxide 
 predominates, and this by contact with the carbonaceous 
 matter becomes reduced to the state of protoxide, and is 
 then not only very fusible, but exercises on all argillaceous 
 matters a very corrosive action, so that crucibles are very 
 seriously injured, and the sides of the furnace require 
 frequent repairs. 
 
 Weight for weight, coke and charcoal give out nearly 
 the same quantity of heat ; but in equal bulks, the former 
 develops much more heat, because its density is greater. 
 From this difference in the calorific power of coke and 
 charcoal, it results that in the same furnace the former 
 produces a greater degree of heat than the latter ; and at 
 high temperatures the difference has been proved to be 
 nearly 10 per cent. In order to account for this, we must 
 consider, firstly, that in a given space the quantity of heat 
 produced in a certain time (and, in consequence, the tem- 
 perature) depends on the amount of fuel burnt, and in- 
 creases with its weight ; secondly, that combustion takes 
 place but at the surface of the masses, whatever may be 
 the nature of the fuel ; from which may be deduced that 
 the weight of fuel burnt in an unit of time ought to be 
 exactly proportionate to its density ; and in consequence, 
 the densest fuels, furnishing the most food for combustion, 
 ought to give out the greatest heat. But, as for the same 
 reason they consume a larger proportion of oxygen, they 
 require, in order to produce the maximum effect, a more 
 rapid and stronger current of air. 
 
 It is clear, from what has been stated concerning the 
 relative properties of coke and charcoal, that when the 
 former can be procured of good quality, and especially 
 when the ash contains but little oxide of iron, it ought to 
 
72 EFFECTS OF FURNACES. 
 
 be preferred to charcoal, for assays requiring a high tem- 
 perature. 
 
 This being an important subject, it has been thought 
 advisable to devote a special chapter to the assay of 
 fuel. 
 
 A very essential condition in obtaining the maximum 
 effect of a furnace, the importance of which can alone be 
 appreciated by experience, is to choose pieces of fuel of a 
 suitable size. If, on the one hand, a shovelful of coke or 
 charcoal be taken at random, it generally contains the 
 dust and dirt found in most fuel, and which, by filling the 
 interstices, prevent the air from passing as required, and 
 consequently render the combustion slow. On the other 
 hand, if a furnace be filled with large pieces, considerable 
 spaces are left between them, so that but a comparatively 
 small surface is exposed to the action of the atmospheric 
 oxygen, and a correspondingly small quantity of fuel is 
 consumed in a given time ; so that the maximum heat 
 can never be obtained. In order to produce the desired 
 result, it is necessary that the pieces shall have a certain 
 mean size, and experience has proved that pieces about 
 1 inch to H inch diameter produce the best effect. 
 These may be selected by sifting the coke through two 
 strong wire sieves, one of which has meshes about 1 J 
 inch square, and the other about 1 inch square. The 
 coke which passes through the larger one, but will not 
 go through the smaller sieve, will be the right size for the 
 furnace. 
 
 THE EFFECTS PRODUCED BY WIND AND BLAST FURNACES. 
 Assays by the dry way can be made either in wind or 
 blast furnaces. In either of them, the degree of heat 
 depends upon the volume of air which passes through 
 the fuel in the same time ; but, cceteris paribus, large 
 furnaces produce more heat than small ones, because 
 comparatively less heat is lost by radiation in the former 
 than the latter. 
 
 In a wind furnace, the maximum of heat is limited by 
 the size of the chimney, and in a blast furnace by the 
 dimensions of the bellows ; but by weighting the latter, 
 
FOCUS OF MAXIMUM TEMPERATURE. 73 
 
 more or less, the force of the blast and, in consequence, the 
 temperature can be increased, to a considerable extent. In 
 this respect blast have the advantage over wind furnaces. 
 
 In the latter, the draught increases in proportion as 
 the heat becomes more intense in the furnace, so that 
 the temperature producible increases progressively. In a 
 blast furnace, the bellows can be weighted and worked 
 as heavily as possible at once, and, by opening all the 
 apertures for receiving air, the maximum temperature can 
 be produced more rapidly than in a wind furnace ; but 
 this is of little use, because, as heat passes very slowly 
 through the substance of a crucible, when the object is 
 to fuse its contents it must be heated gradually, so as 
 to avoid running the risk of softening the crucible before 
 its contents are acted upon, or even scarcely made warm. 
 
 Wind furnaces are, however, much more serviceable 
 and economical than blast, because they work themselves, 
 and do not require the service of a man to attend to the 
 bellows. A blast furnace is used in a laboratory, in cer- 
 tain cases ; for instance, when a single crucible has to be 
 submitted to an intense heat, and when the furnace is 
 small, and the bellows large, in which case the operation 
 resembles a blow-pipe assay. 
 
 In whatever manner the air is introduced into any 
 kind of furnace, whether wind or blast, it is evident that 
 the quantity of heat developed in equal-sized furnaces 
 depends upon the quantity of air introduced in the same 
 time ; but the degree of temperature is not the same in 
 different parts of the furnace, and the distribution of heat 
 varies according to the manner in which the air is intro- 
 duced into the midst of the fuel. The side over which 
 the air passes is kept cold by the current, on which 
 account fire-bars last a long time without becoming oxi- 
 dised, but the heat rapidly augments up to a certain 
 distance from the bars, at which place it arrives at its 
 maximum ; above that it diminishes rapidly, because the 
 air is nearly deprived of its oxygen. Experiment has 
 proved that this maximum is about 2J to 3 inches above 
 the bars or tuyeres. 
 
74 OIL AND GAS FURNACES. 
 
 In common wind furnaces the air enters through the 
 spaces between the horizontal bars which form the bottom 
 of the furnace, and the crucibles are placed on a stand 
 which rests on these bars. By this means the lower and 
 centre part of the crucibles, in which parts the matter to 
 be fused is placed, are exactly situated in the maximum 
 of heat ; but the stand being constantly kept cold, by the 
 contact of a current of air, establishes a continual draining 
 or carrying away of heat from the interior of the crucibles 
 outwards, so that the substance submitted to assay can 
 only arrive at the maximum temperature after a length of 
 time, and the maximum then is always inferior to that in 
 the mass of fuel. It is on this account that assays in a 
 blast or wind furnace generally occupy from one hour to 
 two hours. The author has found that the time may be 
 reduced to half that just stated, if a good solid foundation 
 of fuel be made, and the crucible placed on that, and well 
 surrounded by coke, constantly kept close to the pot and 
 the sides of the furnace ; in this manner the cooling effect 
 of the stand is removed, and the consequent maximum 
 effect of the furnace is produced, but then there is danger 
 of the supporting fuel being burnt away from the crucible 
 and the latter getting upset. 
 
 OIL AND GAS BLAST FURNACES. 
 
 It sometimes happens that metallurgists and assay era 
 have occasion to melt metals at a white heat, but do not 
 wish to heat a large furnace for the purpose. In these 
 cases either the gas or oil furnaces now to be described 
 will prove very useful. 
 
 OIL FURNACES. Mr. Charles Griffin, the well-known 
 chemical instrument maker, has described an oil lamp, 
 which is not only as powerful in action as the best gas 
 furnaces, but almost rivals them in handiness and 
 economy. 
 
 DESCRIPTION OF THE APPARATUS. The oil-lamp furnace 
 is represented in perspective by fig. 32, and in section by 
 fis. 33. It consists of a wick-holder, an oil-reservoir, and 
 
GRIFFINS OIL BLAST FURNACE. 
 
 75 
 
 a fire-clay furnace ; to these must be added a blowing- 
 machine for the supply of atmospheric air. 
 
 The oil-reservoir is represented at a, fig. 32 ; it is 
 made of japanned tin-plate, mounted on iron legs, and 
 fitted with a brass stop-cock and delivery-tube. Its 
 capacity is a little more than a quart. The wick-holder 
 is represented at b, and the upper surface of it by the 
 separate figure <?, fig. 34. The wick-holder and the oil- 
 reservoir are consequently detached, d is a tube which 
 brings oil from the funnel 0, and / is a tube to be placed 
 
 FIG. 32. 
 
 in connection with the blowing-apparatus. The wick- 
 holder contains three concentric wicks, placed round the 
 multiple blow-pipe c, which is in communication with the 
 blowing-tube. 
 
 The crucible furnace consists of the following parts y 
 shown in figs. 32 and 33 : g is an iron tripod ; h is a flue 
 for collecting and directing the flame. The flue is of such 
 a width that when the wick-holder, &, is pushed up into 
 it until the top of the wick is level with the top of the 
 clay cone, there remains a clear air-space of about |- inch 
 all round between the wick-holder and the cylindrical 
 walls of the flue, i represents a fire-clay grate having 
 
76 
 
 DETAILS OF THE MANAGEMENT OF THE 
 
 three tongues, shown by i (fig. 34), on its upper surface. 
 These tongues support the crucible, without stopping the 
 rising flame, k is a fire-clay cylinder which rests upon 
 the grate i, and incloses the crucible, forming, in fact, the 
 body of the furnace. Of this piece there are three sizes : 
 the smallest is of 3 inches bore, and works with crucibles 
 that do not exceed 2f inches diameter ; a middle size, 
 4 inches bore, for crucibles not exceeding 3| inches 
 diameter; the largest size, 5 inches bore, for crucibles 
 not exceeding 4f inches diameter. This piece, being 
 heavy, is provided with handles, as represented at p, 
 
 FIG. 33. 
 
 FIG. 34. 
 
 fig. 34. The walls of the cylinders are from 1 inch to 
 1-| inch thick. I is a fiat plate of fire-clay with a hole in 
 the centre, used to cover the cylinder k, so as to act like 
 a reverberatory dome ; m is a cover which prevents loss 
 of heat from the crucible by radiation, but gives egress to 
 the gaseous products of the combustion of the oil ; n is an 
 extinguisher to put over the wick-holder when an opera- 
 tion is ended ; and o is a support for the wick-holder. No 
 chimney is required. 
 
 MANAGEMENT OF THE OIL-LAMP FURNACE. The apparatus 
 is to be arranged for use as it is represented by fig. 33. 
 The cylinder, , is to be selected to fit the crucible, and 
 
OIL BLAST FURNACE. 77 
 
 the crucible of a size to suit the quantity of metal that is 
 to be melted : 1 Ib. of iron requires the smallest of the 
 three cylinders, described above ; 1^ Ib. the middle size ; 
 5 Ibs. the largest size. The air-way between the crucible 
 and the inner walls of the cylinder should never exceed 
 J inch nor be less than -|- inch. 
 
 The cotton wicks must be clean, and be trimmed a 
 little below the level of the blowpipe c. If properly 
 managed, they do not readily burn away, but can be used 
 for several fusions. The reservoir should be filled with 
 oil for each operation. The proper sort of oil for use is 
 the more volatile kind of mineral oil, of the specific 
 gravity of *750, which is now easily procurable at about 
 three shillings per gallon. The variety known by the 
 commercial name of turpenzine answers well. The com- 
 bustion of a quart of this oil, costing ninepence, gives 
 heat sufficient to melt 5 Ibs. of cast iron. Probably the 
 lighter kinds of paraffin oil may be suitable. Liquids of 
 the alcoholic class, spirits of wine, and pyroxylic spirit 
 can be -used ; but they are less effective and more expen- 
 sive than turpenzine. Care must be taken not to spill 
 the oil on the table or floor, and not to decant it carelessly 
 in the neighbourhood of a light, because atmospheric air 
 strongly charged with the vapour of these light oils is 
 explosive. -When the oil is burnt in the furnace in the 
 manner described below, there is no danger. During an 
 operation, a wooden screen, as represented by the dotted 
 lines in fig. 32, should be placed between the oil-reservoir 
 and the furnace, to prevent the vaporisation of the oil by 
 radiant heat. As the wick-holder b, and the supply pipe 
 d) contain only about one fluid ounce of oil, the oil must 
 run continuously during a fusion, from the reservoir #, 
 into the funnel 0, in order that the cotton may be always 
 flooded. The success of the fusion depends upon the due 
 supply of oil, to which point the operator must pay atten- 
 tion. At the commencement of a fusion, the oil must be 
 run from the reservoir until the surface of the oil in the 
 funnel has a diameter of about an inch. The wicks will 
 then be flooded, and a lidit may be applied, and a gentle 
 
78 THE OIL BLAST FURNACE. 
 
 blast of air then turned on. The oil immediately sinks in 
 the funnel, and the stop-cock must be opened and so regu- 
 lated as to keep the oil barely visible at the bottom of the 
 funnel. If too much oil is supplied it immediately rises 
 in the funnel, and simultaneously overflows the wick- 
 holder. Too much vapour is then thrown into the fur- 
 nace, and the heat is immediately lowered, especially at 
 the beginning of an operation, before the fire-clay portions 
 of the furnace are well heated. If, on the contrary, too 
 little oil is supplied, the wicks burn, and the operation is 
 spoilt. The demand of the wick-holder for oil depends 
 upon the condition of the furnace and the character of the 
 fusion in progress. When the lamp is newly lighted and 
 the furnace cold, the oil should be passed slowly in distinct 
 drops ; but as the furnace becomes heated the rapidity of 
 the supply of drops should be increased ; and finally, when 
 the furnace is at a white heat, the oil should be supplied 
 in a thin continuous stream. When the fusion to be 
 effected is that of only a small quantity of metal, such as 
 1 Ib. of iron, a rapid supply of drops of oil is sufficient 
 even to the close of the operation. At that rate the 
 burner consumes about 1 J pint of oil in an hour. When 
 the fusion to be effected is that of 4 Ibs. or 5 Ibs. of iron 
 and the large furnace is in action and has been brought 
 to a white heat, the supply of oil must, as stated above, be 
 in a thin continuous stream, and the operation will then 
 consume two pints of oil in an hour. And here it requires 
 remark that, with that continuous supply, when the fur- 
 nace is large and is at a white heat, the oil does not rise 
 in the funnel, being instantaneously converted into gas at 
 the mouth of the burner, and thrown up in that state into 
 the furnace for combustion. The operation, indeed, con- 
 sists at that point of a rapid distillation of oil-gas, which 
 is immediately burnt, in the presence of air supplied at a 
 suitable pressure by a dozen blowpipes, in effective contact 
 with the crucible to be heated. 
 
 The flame produced in this furnace is as clear as that 
 produced by an explosive mixture of air and coal-gas. It 
 is perfectly free from smoke, and the consumed vapours 
 
POWER OF THE FURNACE. 79 
 
 which occasionally escape with gaseous products of a com- 
 bustion are even less unpleasant to smell and to breathe 
 in than are those which are usually disengaged by a blast 
 gas furnace, or by an ordinary lamp, fed with pyroxylic 
 spirit. 
 
 The contents of a crucible under ignition in this furnace 
 can at any moment be readily examined, it being only 
 necessary to remove the pieces I and m with tongs, and 
 to lift the cover of the crucible, during which the action 
 of the furnace need not be interrupted. 
 
 When the operation is finished, the blast is stopped, 
 the stop-cock is turned off, the oil-reservoir is removed, 
 the wick-holder is lowered on the support 0, withdrawn 
 from the furnace, and covered with the extinguisher n. 
 The quantity of oil which then remains in the lamp is about 
 one fluid ounce. 
 
 POWER OF THE OIL-LAMP FURNACE. The furnace being 
 cold when an operation is commenced, it will melt 1 Ib. of 
 cast iron in 25 minutes, 1^ Ib. in 30 minutes, 4 Ibs. in 45 
 minutes, and 5 Ibs. in 60 minutes. When the furnace is 
 hot, such fusions can be effected in much less time ; for 
 example, 1 Ib. of iron in 15 minutes. It need scarcely be 
 added that small quantities of gold, silver, copper, brass, 
 German silver, &c., can be melted with great ease, and 
 that all the metallurgical and chemical processes that are 
 commonly effected in platinum and porcelain crucibles can 
 be promptly accomplished in the smallest cylinder of this 
 furnace ; and in the case of platinum vessels, with this 
 special advantage, that the oil-gas is free from those sul- 
 phurous compounds the presence of which in coal-gas fre- 
 quently causes damage to the crucibles. 
 
 REQUISITE BLOWING POWER. The size of the blowing 
 machine required to develop the fusing power of this oil- 
 lamp furnace depends upon the amount of heat required 
 or the weight of metal that is to be fused. For ordin- 
 ary chemical operations with platinum and porcelain 
 crucibles, and even for the fusion of 1 Ib. of cast iron in 
 clay or plumbago crucibles, a blowing power equal to that 
 of a glass-blower's table is sufficient, provided the blast it 
 
80 GAS FURNACE. 
 
 gives is uniform and constant. But the fusion of masses 
 of iron weighing 4 or 5 Ibs. can be effected by the gas 
 which this oil-lamp is capable of supplying, provided a 
 sufficiently powerful blowing machine supplies the requi- 
 site quantity of air. When more than a quart of oil is to 
 be rapidly distilled into gas, and the whole of that gas is 
 to be burned with oxygen, it is evident that effective work 
 demands a large and prompt supply of air. 
 
 As in all practical matters of this sort the cost is a 
 main question, it may be useful to state that the price of 
 this apparatus complete, without the blowing machine, but 
 including every other portion necessary for heating cruci- 
 bles up to the size sufficient to fuse 1 Ib. of cast iron, is one 
 guinea ; and that with the extra furnace-pieces for cruci 
 bles suitable for 5 Ibs. of iron, or any intermediate quantity, 
 the cost is one guinea and a half. 
 
 Mr. Griffin has described before the Chemical Society 
 an improved gas furnace for chemical operations at a white 
 heat without the aid of a blowing machine, and a new 
 method of supporting crucibles in gas furnaces. The fol- 
 lowing extracts from his paper are taken from the Journal 
 of the Chemical Society : 
 
 4 On a former occasion I introduced to the notice of 
 the Chemical Society a gas furnace for operations at a 
 white heat in crucibles, or a copper-melting heat in muffles. 
 A detailed description of that furnace is given in the 
 Journal of the Society for August 1870. The crucibles 
 are either suspended in a pierced plumbago cylinder or 
 supported on a trivet grate, both of which are liable 
 to break when white hot, and therefore cause trouble 
 and expense. Crucibles vary so much in form and size 
 that they are often not suspended from these cylinders 
 exactly in the focus of the heating power. Trivet grates 
 have the disadvantage that they interfere with the direct 
 action of the flame upon the crucible, and if made slightly 
 they break when heated to whiteness. I desire now to 
 place before you a new form of burner by which these 
 defects are remedied. In the new burner the circle of gas- 
 jets is enlarged so as to leave a space round the central 
 
GAS FUKNACE. 81 
 
 jet. An atmopyre similar to those used in Hofmann's 
 combustion furnace, but of greater bulk and strength, is 
 dropped over this central jet, and forms a solid support 
 for the crucible. This support does not readily break, 
 but, should an accident happen, it can be replaced at the 
 cost of a few pence. It brings the bottom of the crucible 
 exactly into the focus of heat, and itself supplies a portion 
 of the heating power of the burner. It also enables one to 
 use any crucible at hand, independent of its form or size. 
 A strong lateral arm cast on the body of the burner sup- 
 ports an upright iron rod, which carries the chimney of 
 the furnace. By prolonging the legs of the burner up- 
 wards they are made to carry the clay furnace, and thus, 
 by doing away with a stool or other support, the con- 
 struction is simplified and the cost lessened. A plumbago 
 cylinder, to deflect the flame and entrap the heat, is placed 
 round the crucible, and is covered with an ordinary 
 crucible cover, by removing which the crucible can be 
 inspected. These fittings, however, present nothing new, 
 being adapted from Griffin's gas blast furnace, which was 
 introduced sixteen years ago. Access to the crucible in 
 the furnace is gained by turning aside the chimney and 
 lifting the top plate of the furnace, which is provided with 
 handles for this purpose. These handles do not become 
 very hot, even when the furnace is at a white heat. The 
 power of these new burners is very remarkable, one of 
 small size consuming only 20 feet of gas per hour, and 
 having a chimney 4 feet high, being capable of fusing half 
 a pound of cast iron in 35 minutes from the time of lighting 
 the gas ; or of melting gold, silver, or copper in crucibles 
 placed within a muffle measuring 5 inches long by 3 wide. 
 If a chimney 6 feet high be employed, cast iron can be 
 melted in crucibles placed within the muffle. A burner of 
 larger size, consuming 40 feet of gas per hour, will melt 
 cast iron in crucibles placed within a large muffle measuring 
 8 inches long by 4 inches wide. In the crucible furnace 
 it will melt 1 Ib. of cast-iron in 35 minutes, 2 Ibs. in 
 45 minutes, 3 Ibs. in 55 minutes, and 4 Ibs. in 65 
 minutes, from the time of lighting the gas. It is thus 
 
 G 
 
82 
 
 GAS FURNACE. 
 
 seen that, when a white heat has been once obtained, 
 10 minutes' time is required for the fusion of every 
 additional pound of iron. These results, attainable with 
 certainty and rapidity, are, I believe, the highest that have 
 hitherto been placed at the command of the chemist. As in 
 my former furnace, the proper admixture of gas and air 
 is judged of from the colour and quantity of flame which 
 passes up the chimney. To enable the operator to see 
 this flame, three small holes are bored in the chimney. 
 
 FIG. 35. 
 
 FIG. 36. 
 
 The flame is not seen at the upper hole, unless the supply 
 of gas is too large, but it is always visible at both the 
 lower holes.' 
 
 In the above figure the muffle is provided with a small 
 draught flue, having a regulating cap on its upper end. 
 In the small furnaces this is omitted, and the muffle is 
 slotted in the usual manner. The cover of the furnace is 
 now made without the zigzag opening in the roof. The 
 burner of the muffle furnace is the same as that used in 
 the crucible furnace, fig. 35. 
 
GAS FURNACE. 83 
 
 Skittle pots up to 8 inches can be used in these furnaces 
 for collecting and burning waste with fluxes, &c., and in 
 much lees time than is required by a coke fire, An 8-inch 
 pot can be worked in half an hour from lighting the gas. 
 Two of the outer cylinders are used, placed one on the 
 top of the other. 
 
 FIG. 37. 
 
 With a 4-foot flue the muffle furnace melts gold, silver, 
 and copper ; with a 6-foot flue it melts cast iron placed i 
 crucibles within the muffle. The consumption of gas is 20 
 cubic feet per hour. 
 
84 UNIVERSAL GAS FURNACE. 
 
 The Distillation of Pure Zinc, per descenswn, can be 
 performed in one of these gas furnaces by passing a tube 
 from the top of the crucible downwards through the 
 burner to the table. 
 
 Fig. 37 shows one of Griffin's extra large gas fur- 
 naces, capable of raising a No. 12 plumbago pot, measur- 
 ing 8 inches high by 6 inches wide, to a white heat. The 
 cover of this furnace is let into the body, which rises higher 
 .than in the smaller patterns, and from which the flue passes 
 off laterally to a standing flue or other house chimney. 
 
 UNIVERSAL GAS FURNACE. Mr. Thomas Fletcher has 
 devised what he terms a universal gas furnace ; it works 
 without blast for crucible operations, up to a clear white 
 heat. 
 
 The specialty of this furnace is the burner. It is as 
 simple and easy to use as an ordinary Bunsen's burner, but 
 the flame is solid to the centre, unlike the flame of every 
 heating burner which has previously been made. The 
 open flame will readily fuse a coil of thick copper wire, 
 and to make a crucible furnace it simply requires a support 
 for the crucible, and a fire-clay jacket to prevent radia- 
 tion, as the flame is in itself perfect, and requires no 
 blowing or attention in any way. The furnace is so small 
 and light that it can be used on the work-bench, and put 
 away on a shelf, and can be used on a sitting-room table 
 without the slightest dirt or inconvenience. The body of 
 the furnace is only 6 inches in diameter. 
 
 THE SINGLE-JACKETED ARRANGEMENT is capable of melting 
 5 or 6 oz. of gold in 15 minutes with a 10-inch chimney, 
 or in 10 minutes with a 2-ft. chimney. With the ladle- 
 holder it will melt 8 Ibs. of zinc in 15 minutes, without 
 chimney, in an ordinary iron ladle, and lead, tin, &c., in 
 a proportionately shorter time. 
 
 THE DOUBLE-JACKETED CHEMICAL FURNACE, having the same 
 burner as above, and requiring no more gas, will melt 3 
 or 4 oz. of c.ast iron in 35 minutes, if used with a 3-ft. 
 chimney, or, if with a longer chimney, in a proportion- 
 ately shorter time, and will give any required tempera- 
 ture in proportion to the length of chimney used, provided 
 
FLETCHER'S GAS FURNACE. 
 
 the gas is turned on sufficiently just to cover the crucible 
 with flame when the chimney is in its place. 
 
 ' In case of the fusion of a crucible, or its penetration 
 by fluxes when the furnace is used for extremely high 
 temperatures, no damage can be done to the furnace, 
 except perhaps the destruction of a few of the burner 
 tubes, which can be replaced at a trifling expense. No 
 part of the furnace is liable to injury with constant 
 use. 
 
 The lower part of the burner is a chamber 6 in. x 
 3 in., open at the bottom, in which the gas is partially 
 mixed with air. This mixture is conducted to the top of 
 the burner through a mass of fine tubes,. 55 in number, 
 with an arrangement to supply between each exactly the 
 amount of air necessary to consume it instantly. A flame 
 produced by this means, consuming 20 feet of gas per 
 hour, is only about 2 inches high, and almost colourless. 
 The whole of the available heat is generated in the proper- 
 place, viz. below the object to be heated, which, there- 
 fore, is not also cooled by the passage of unburnt gas and 
 air. The flame is very similar in appearance to a mass of 
 blowpipe jets, and has the same heating power. The point 
 of greatest heat commences, as with a blowpipe, at the 
 point of the blue cones, about ^ in. or f in. above the 
 tubes. 
 
 The Double-jacketed Chemical Furnace is made to 
 take crucibles not exceeding in size No. 00 of the Plum- 
 bago Crucible Co., which are the most convenient for 
 melting quantities not exceeding 5 or 6 oz. of gold, &c. 
 
 The gas supply tap and pipe must be large and clear, 
 so as to give as great a pressure of gas as possible at the 
 burner nozzle, although the actual consumption of gas is 
 small. The indiarubber tubing used must of necessity 
 be perfectly smooth inside. The tubing made on wire, 
 whether the wire is removed or not, will not work these 
 burners satisfactorily. The gas supply specified is re- 
 quired to work each furnace at its full power, and the 
 flame must be visible in the chimney. If the gas supply 
 is deficient, the furnaces can be worked at a lower heat 
 
o FLETCHER'S GAS FURNACE. 
 
 by partially closing the top of the chimney until the flame 
 becomes visible, or by working without the chimney. If 
 the burner plate becomes red hot, it is a sign that the 
 gas supply is deficient. If the supply of gas is not suffi- 
 cient, it will be found necessary to examine the tap which 
 supplies the gas, many of which, as supplied by gasfitters, 
 are exceedingly faulty, and partially stopped up. If this 
 be the case, the tap should be replaced by a better one. 
 The furnace requires supply from a f or T 7 in. pipe. If 
 a larger crucible be used than the one supplied, an extra 
 fire-clay ring must, be obtained of the same proportionate 
 width, and half an inch higher than the top of the cru- 
 cible. 
 
 Where the gas supply is more than necessary, a longer 
 chimney may be used with advantage, but, as the furnace 
 itself is too small and light to form a steady support for a 
 longer chimney, it will be necessary to suspend the upper 
 part with a wire, or to support it with a bracket from 
 the wall. The furnace requires the following supply 
 of gas : Without chimney, 18 cubic feet ; with 10-inch 
 do., 22 cubic ft. ; with 36-inch do., 28 to 30 cubic ft. per 
 hour. 
 
 When a white heat is required, the crucible must be 
 covered with an inner perforated plumbago dome, and 
 which forms the inner jacket, with the perforations in 
 such a position that the crucible can be seen through the 
 hole in the lower part of the chimney ; a chimney not 
 less than 3 feet high must be used. If a greater heating 
 power is required, a longer chimney will enable it to be 
 obtained in the same or a shorter time, but under all 
 circumstances the flame must just cover the crucible to 
 obtain proper results. 
 
 The addition of a small muffle for high temperatures 
 makes this furnace complete for all purposes. The clear 
 working space, which is equally heated in every part, 
 measures 2^ x 2f x 2J in. The muffle is not in contact 
 with the outer casing in any part, and therefore all parts 
 are equally heated. The temperature obtained depends 
 on the length of chimney used, provided the gas supply is 
 
FLETCHER'S GAS FURNACE. 87 
 
 sufficient to cover the muffle with flame when the chimney 
 is on. For silver assays about 4 feet of chimney should 
 be used, which gives the melting point of fine silver in 16 
 minutes from the time the gas is lighted; the same tem- 
 perature being always obtained in exactly the same time, 
 within a few seconds ; and after the first trial the opera- 
 tion does not require watching. With this furnace there 
 will be found no variation in a number of assays done at 
 different times. 
 
 A chimney of about 6 or 8 ft. gives nearly a clear 
 white heat in 30 minutes, and for special operations 
 requiring very high temperatures a longer chimney may 
 be used as required. A chimney or stove pipe 10 or 12 
 ft. long may be used as a fixture, and the draught par- 
 tially stopped with a damper or slide when lower tempe- 
 ratures are required, the gas being turned down in pro- 
 portion ; the guide for proper adjustment being, that 
 under all circumstances the flame must just cover the 
 muffle, but must not extend into the chimney so as to 
 make it red hot. 
 
 It will not be found necessary in practice to partially 
 close the chimney for lower temperatures, as the same 
 effect may be produced equally well by simply turning the 
 gas lower ; but it is more economical to partially stop the 
 draught, so as to prevent excess of cold air being drawn 
 in. When the flame covers the muffle, the gas is doing 
 its extreme duty under the most favourable circumstances, 
 without waste. The same rule applies also to the crucible 
 arrangement. Particles of flux should not be allowed to 
 fall on the fire-clay casing, where the parts touch each 
 other ; and the power of the furnace should not be urged 
 too far by the use of very long chimneys, as there is 
 danger of the fusion of the fire-clay parts together so that 
 they cannot be separated. Fire-clay fittings, as a rule, 
 cannot be safely used for temperatures much exceeding 
 the fusing point of cast iron. 
 
 An excellent gas assay furnace, slightly modified from 
 one of Mr. Fletcher's, has been described in the Chemical 
 News by Mr. Walter Lee Brown, and is shown in fig. 38. 
 
88 FLETCHER'S GAS FURNACE. 
 
 As shown, the form is almost that of the reverberatory 
 furnace, the movable bricks when in place forming the 
 roof. From another point of view, it may be described 
 as the muffle of an ordinary furnace, but having the flame 
 as well as the heat inside. The exterior dimensions are 
 as follows : 20 inches long, 7 inches wide, and 5i inches 
 deep. The nozzle of the burner is connected with a 
 | inch tap by means of a full ^ inch rubber tube. A 
 3-inch stove-pipe, tightly fitting, is connected with a flue. 
 
 In the interior, upon the floor, rest four little wedge- 
 shaped pieces of fire-clay which are movable, and upon 
 
 FIG. 38. 
 
 them rests a false floor, also movable. The latter (not 
 shown in the cut) corresponds to the mutfle bottom of an 
 ordinary furnace, and upon it is done all the work. It 
 is 3^ inches wide by 7^ inches long, and will accommo- 
 date three 2f inch or four 2J inch scorifiers, or eighteen 
 1 J inch cupels at once ; but like any other furnace, it is 
 better not to crowd it. 
 
 The manner of operation is simple. The covering 
 bricks are removed, the milled wheel at the gas entrance 
 to the burner is turned back so as to allow a full flow of 
 gas, the handle at the supply tap turned full on, the 
 gas lighted, and the bricks put into place. The flame, 
 
FLETCHER'S GAS FUKNACE. 89 
 
 when the full amount of gas is used, will be highly car- 
 buretted, and of course strongly reducing, but intensely hot. 
 In from fifteen to twenty minutes the interior will be hot 
 enough for work. The bricks are removed, the charged 
 scorifiers are placed within, the bricks set back, the excess 
 of gas turned off at the milled wheel, and the furnace kept 
 closed till the charges are melted. After this has been 
 effected, the bricks are separated or slid aside more or 
 less to admit of air for scorification. Proper regulation is 
 made by the milled wheel, by which the gas may be 
 turned partially off as required, always leaving the supply 
 tap fully open. The action of the air is also controlled by 
 the damper in the pipe leading to the chimney. 
 
 For cupelling, the gas supply is turned down more than 
 in scorifying. The time of performing either scorification 
 or cupellation varies according to the nature of the ore 
 and other circumstances, but is about the same as in the 
 use of a coke furnace. 
 
 This furnace does well for small crucible fusions, by 
 removing the false floor and its supports. 
 
 The advantages of this furnace are many : convenience 
 of operating whereby the assayer sees every step and 
 stage of the operation, and so can tell when and where to 
 change or improve ; perfect control of the source of heat, 
 so that a higher or lower temperature, a reducing or 
 oxidising effect, may be produced in an instant ; entire 
 noiselessnesss, in which characteristics it is the superior 
 of all blast assay furnaces ; saving of time, which in 
 furnaces employing coke, charcoal, or coal is spent in 
 4 bedding down,' feeding, &c. ; freedom from the annoy- 
 ances of dust and ashes ; absence of waste, for when the 
 work is performed the gas is at once shut off; comfort in 
 manipulation, for it does not heat up a room as do most 
 furnaces (quite a desideratum in the summer time) ; 
 finally, its remaining qualifications, which need not be 
 dwelt upon, are simplicity of construction, durability, and 
 portability. 
 
 The consumption of gas is not far from 30 cubic feet 
 per hour. 
 
90 FLETCHER'S GAS FURNACE. 
 
 Fig. 39 shows one of Fletcher's Eeverberatory Gas Fur- 
 naces for crucibles, muffles, cupels, &c. 
 
 One of these furnaces will do most of the general work 
 of an ordinary laboratory. They work perfectly with 
 chimney draught to a bright red about the fusing point 
 of fine copper and fine silver. With a blast they will 
 work up to the fusing point of cast iron. The furnaces can 
 be made to take either two muffles at once, a number of 
 crucibles, trays of cupels, or one muffle and crucibles or 
 cupels at the same time. 
 
 The opening may be either at the side or the top, the 
 furnace working either way equally well. The burner is 
 
 FIG. 39. 
 
 at one end, out of the way of injury in case of accident 
 to a crucible. Crucibles, cupels, &c., stand on the solid 
 bottom of the furnace, perfectly steady and firm. When 
 a blast burner is used a clay collar fits into the larger 
 opening necessary for a draught burner, and the instruc- 
 tions for both draught and blast for the ordinary furnaces 
 apply" equally well to this, the burners being identical in 
 principal with those of the previous patterns. 
 
 When used with the draught burner the blue cones 
 of flames must be clearly seen on the burner, or if they 
 disappear the gas supply must be increased, or the slide 
 over the burner air tube closed until they reappear. In 
 the latter case the furnace works with a smaller gas supply 
 at a lower temperature, and by closing this slide and 
 
GORES GAS FURNACE. 
 
 91 
 
 reducing the gas supply any temperature required can 
 be obtained. If the adjustment of gas and air is neglected 
 the burner grid becomes red hot and is quickly rendered 
 useless. The grids will last for years if properly used. 
 
 A good 
 
 muffle fur- 
 nace to work with gas 
 has long been a deside- 
 ratum in the assay labo- 
 ratory. Mr. Fletcher 
 has supplied this want 
 in the furnace shown in 
 fig. 40. 
 
 The gas required for 
 this furnace is 70 cubic 
 feet per hour. There 
 
 FIG. 40. 
 
 be 
 
 -| inch 
 
 clear 
 
 the 
 
 gas 
 
 must 
 
 bore through 
 
 pipes and tap. 
 
 Fig. 41 shows one of 
 Fletcher's Draught Crucible Furnaces. This will melt 
 brass, silver, copper, and gold, but is not suitable for cast 
 
 FIG. 41. 
 
 <* It takes about 25 cubic feet of gas per hour, and 
 requires a i inch pipe and tap. The largest size crucible 
 
 iron. 
 
92 
 
 GORES GAS FURNACE. 
 
 it will take is 4x3^ inches, and it will melt 6 Ibs. of 
 brass. 
 
 Mr. Fletcher has recently brought out a new melting 
 arrangement for melting up to 3 oz. of gold or silver, 
 FIG. 42. rapidly, without the use 
 
 of a furnace. 
 
 The figure (42). is 
 slightly under half-size. 
 In this arrangement the 
 two parts of the ingot- 
 inould slide on each other , 
 to enable ingots of any 
 width to be cast, And 
 the blowpipe is part of 
 the rockingr-stand. Con- 
 nect the blower to the 
 upper tube, and the gas 
 to the lower. When the 
 metal is melted in the 
 shallow crucible of com- 
 pressed charcoal, tilt the 
 whole apparatus over so as to fill the ingot-mould. A 
 sound ingot can be obtained in about 2 minutes. Very 
 bulky scrap should be run into a mass in one of the 
 moulded carbon blocks before being placed in the cru- 
 cible. No flux must be used with the charcoal crucibles. 
 With a larger- sized melting arrangement similar to the 
 above, as much as 14 ounces of fine silver, or 20 ounces 
 18-carat gold, can be melted and cast in an ingot in 5 or 6 
 minutes. This size requires a ^ inch gas supply and a foot- 
 blower. 
 
 Mr. Fletcher has also devised an injector gas furnace 
 (fig. 43) for metallurgists, jewellers, chemists, iron, brass, 
 and nickel founders, manufacturers of artificial gems, and 
 other purposes where an ordinary furnace is useless or 
 unreliable. 
 
 It has been found that in working at extremely high 
 temperatures, the ring which holds the gauze is liable 
 to be fused. To prevent this, a new burner has been 
 
GORE'S GAS FURNACE. 90 
 
 designed, in which the ring is entirely dispensed with, and 
 the gauze cap is pushed up from the back of the burner 
 against a small shoulder inside the nozzle of the burner. 
 The burner is in one 
 
 casting, and therefore FlG - 43 - 
 
 there is no tendency 
 for the nozzle to get- 
 hot, as in the former 
 pattern. See that the 
 gauze is pushed up AIR CHECK 
 from behind to within AIR 
 about J inch of the 
 nozzle. The power 
 and speed of working 
 are practically without limit, depending only on the gas 
 and air supply, and are under perfect control. With 
 ^ inch gas pipe and the smallest foot-blower, the small 
 furnace will melt .a crucible full of cast-iron scrap in 7 
 minutes, and steel in 12 minutes, starting with all cold. 
 
 To adjust a new furnace to its highest power, put 
 the nozzle of the burner tight up against the hole in 
 the side of the casing, turn on the full gas supply, 
 with the air-way full open. If the flame comes out of 
 the lid about 2 inches, the adjustment is right. If the 
 flame is longer, enlarge the hole in the air-jet until the 
 proper flame is obtained, or reduce the gas- supply ; if 
 smaller, or not visible, turn the air-check until the flame 
 appears. 
 
 Keep all fluxes away from the furnace jacket. Before 
 stopping the blower, draw the burner back from the 
 hole. Commence blowing before the lid is put on the 
 furnace. 
 
 The old pattern blower is liable to pick up dirt from 
 the floor, throwing it against the gauze of the burner, and 
 stopping the proper working of the furnace until cleared 
 away. A thin layer of silver sand on the bottom will 
 prevent crucibles adhering when at a white or blue heat. 
 Crucibles must be heated very slowly the first time they 
 are used, unless of the ' Salamander ' brand. 
 
94 
 
 GORE S GAS FURNACE. 
 
 In cases where gas cannot be obtained, Mr. Fletcher has 
 devised a simple furnace (fig. 44) for high temperatures, 
 
 working with either 
 FIG. 44. fo . ., 
 
 gas or spirit-petroleum 
 
 without alteration, and 
 giving perfect results 
 with either fuel. This 
 furnace is supplied 
 with a small, simple, 
 and perfectly safe ar- 
 rangement for burning 
 the vapour of gasoline 
 or benzoline, giving a 
 power and efficiency 
 fully equal to that 
 which can be obtained 
 by a larger gas supply. 
 The apparatus is in 
 
 every way as simple as- when gas is used, requiring no more 
 trouble or attention. ' 
 
 It equals a gas furnace in every respect, and, in 
 addition, gives a heat of absolute purity, fitting it for the 
 most delicate chemical operations where gas cannot be 
 used owing to the presence of sulphur and other matters. 
 The ordinary pattern of injector furnace is used in 
 precisely the same way as with gas, the only difference 
 being that a branch pipe is taken out of the air supply 
 and connected to the lower tap A on the generator, and a 
 tube is carried from the upper tap B to the side tube of 
 the injector burner marked 'gas.' The quantity of vapour 
 required is adjusted by the lower tap A when the furnace 
 is working, and the flame must be just visible at the hole 
 in the lid, exactly as when gas is used, the instructions 
 being precisely the same 'for both fuels. 
 
 To charge the generator, pour benzoline or gasoline in 
 the top hole until it overflows at the small tap C in the 
 side, replace the cork firmly, and close the overflow tap. 
 It will then work for about 10 to 12 hours at the full . 
 power of the furnace. ' ' ! o 1 
 
GORES GAS FURNACE. 95 
 
 Benzoline varies much in quality ; it must, when a 
 few drops are poured on a plate or the hand, evaporate 
 quickly and completely, leaving no greasy stain, and if 
 good will produce more vapour than the furnace can burn 
 at its maximum power. All the tubing used must be 
 perfectly smooth inside, or the power of the furnace will 
 be greatly reduced. 
 
 At the conclusion of an operation, close both taps on 
 the generator. It can then be left for any length of time 
 ready for instant use. For ordinary meltings, the gene- 
 rator can be used about thirty or forty times without 
 refilling. 
 
 The No. 3 size will refine and perfectly fuse 6 Ibs. of 
 chemically pure nickel so that it can be poured clean, 
 using an open crucible, a feat beyond the capabilities of 
 any other known furnace. 
 
 Benzoline often contains heavy oils. If the generator 
 works badly, empty it and refill with fresh. 
 
 G. Gore, Esq., F.E.S., has devised a gas furnace which 
 will fuse cast iron, &c., and still allow the melted sub- 
 stances to be perfectly accessible to be manipulated upon 
 for a continuous and lengthened period of time, without 
 contact with impurities or with the atmosphere, and with- 
 out lowering their temperature sufficient to cause them 
 to solidify. These conditions Mr. Gore has obtained by 
 means of ordinary coal-gas and atmospheric air, without 
 the use of a bellows or lofty chimney, or of regenerators 
 or valves requiring frequent attention. The arrangement 
 is as follows : A (figs. 45 and 46) is a cylinder of fire-clay 
 about 9 inches high and 6 inches diameter, open at both 
 ends, with a hole in its side near the bottom to lead into 
 the chimney ; it is covered by a movable plate of fire-clay, 
 B, with a hole in its centre for the introduction or removal 
 of the crucible, c. ; this hole is closed by a perforated plug 
 of clay, C, for access to the contents of the crucible, and 
 that again is closed by another clay stopper D ; E is a 
 chimney of sheet iron about 5 or 6 feet high, kept 
 upright by a ring of iron F attached to the top of the 
 furnace. The fire-clay cylinder is enclosed in a sheet of 
 
90 
 
 GORES GAS FURNACE. 
 
 iron casing with a bottom of iron, to which are fixed 
 three iron legs G. An iron tube H, with a prolongation 
 /, supports by means of the screw / the burner K and its 
 tube jL, which is open at both ends. Gas is supplied to 
 the burner by means of the tap M, which has a small 
 index A 7 attached to it for assistance in adjusting the gas. 
 Inside the large cylinder is another fire-clay cylinder or 
 cupola 0, with open ends, and with three projections of 
 
 FIG. 45. 
 
 FIG. 46. 
 
 fire-clay P, for supporting the crucible Q ; it is kept 
 steady by means of three clay wedges R ; S is an air-valve 
 for closing the bottom of the tube L. The gas-burner is 
 a thin metal cylinder, deeply corrugated at its upper end, 
 with the corrugations diminishing to nothing at its lower 
 end, as shown in the engravings. The action of this 
 furnace is as follows : Gas is admitted to the open tube L 
 by the tap M ; it there mixes with air to form a nearly 
 combustible mixture, which ascends through the burner, 
 
GORES GAS FURNACE. 97 
 
 and burns in the clay cylinder 0, being supplied with the 
 remainder of air necessary for combustion through the 
 tube H to the outer surface of the flame ; the products of 
 combustion pass up through the cylinder 0, and then 
 downwards outside of it to the chimney, the point of 
 greatest heat being at Q. 
 
 It is important in using this furnace that the burner be 
 placed quite in the centre of the bottom of the tube ; 
 also that a crucible of not too large nor too small dimen- 
 sions be selected. The most suitable way of supporting a 
 smaller crucible is by placing it in a larger one that has 
 had its upper part broken off. If desirable, a little clay 
 luting may be placed round the top edge of the iron 
 casing to exclude air entering between it and the cylinder ; 
 also a little thin clay luting upon the part of the bottom of 
 the furnace where the inner cylinder rests. 
 
 In lighting the furnace, the plugs C and D are removed, 
 a light held inside the opening, and the gas turned on full. 
 Should the flame blow down to the bottom of the tube L 
 on lighting (which, however, rarely occurs unless . the 
 furnace is already hot), the gas must be turned off, and 
 the bottom of L momentarily closed whilst lighting the gas 
 as before. Should the flame not burn down to the burner, 
 but only burn to the orifice in the clay plate B, it must at 
 once be extinguished and relighted, otherwise some of the 
 gaseous mixture will pass into the chimney unburned, and 
 subsequently ignite and cause an explosion. A large 
 flame now issues from the top orifice, and is white if too 
 much gas is on. and chiefly violet or red with the proper 
 quantity ; it should now be coarsely adjusted until these 
 appearances are represented. The annular plug C should 
 now be inserted, which will compel it to pass downwards 
 to the chimney, and as soon as the small remaining flame 
 now issuing disappears, or nearly disappears, as it will in 
 a few seconds, the smaller stopper D should also be 
 inserted. In lieu of this, the large flame may be deflected 
 against the chimney by means of a piece of sheet iron until 
 it withdraws inwards as before mentioned ; the two plugs 
 may then be reinserted. The gas tap may now be partly 
 
 H 
 
98 -GORES GAS FURNACE. 
 
 adjusted. The crucible should be placed in the furnace 
 after the act of lighting the gas, but not immediately after 
 if the furnace is cold, or explosions may occur by un- 
 burned gaseous mixture passing the crucible into the 
 chimney, and igniting afterwards. 
 
 After about five minutes the gas should be slowly ad- 
 justed, until a sound is heard inside like a series of small 
 explosions. This sound is sometimes not very distinct, 
 especially at high temperatures, and therefore requires a 
 little experience in the use of the furnace in order to be 
 detected. It is, however, a chief guide in determining the 
 proper amount of gas, and should therefore be carefully 
 studied. To assist in adjusting the gas it will be found 
 very useful to place a small piece of looking-glass beneath 
 the tube Z, and to adjust the gas tap until the flame 
 between the -burner and crucible appears wholly violet or 
 slightly white ; but this test is liable to fallacy if em- 
 ployed when the gas is first lighted, because the coldness 
 of the parts makes the flame much whiter than it other- 
 wise would appear. It is also fallacious, by the flame 
 appearing whiter than it really is when the crucible is 
 very hot. It is, however, of great assistance, especially at 
 intermediate temperatures. A rough deposit upon the 
 outer edge of the crucible indicates an excess of gas ; the 
 deposit is carbon. Less gas is required with a crucible in 
 the furnace than without one : also less is required when 
 the small hole at the top of the furnace is open than when 
 it is closed ; and less is also required when the furnace is 
 cold than after it has been lighted some time, because the 
 draught gradually increases and draws in more air. After 
 having accurately adjusted the gas, no further attention 
 to the furnace is requisite. 
 
 Having once found the proper adjustment of gas under 
 certain known conditions, it is well to notice the position 
 of the index pointer N, in order to be able at once to 
 adjust it to about the right point on other occasions. 
 Under ordinary circumstances, during daylight it is best 
 to set the gas nearly full on at first, and fully on at about 
 five minutes afterwards when the draught has become more 
 
 
GORE'S GAS FURNACE. 99 
 
 powerful ; but during twilight, when the supply of gas 
 from the gas works is more free, the index pointer may be 
 set at the numbers 2-J or 3. The gas should be supplied 
 by a pipe of not less than f-inch bore, with a main pipe of 
 ^ an inch ; but all depends on the pressure of gas at the 
 particular locality, which is very variable. The consump- 
 tion of gas varies from 30 to 40 cubic feet per hour, the 
 value of which is about twopence. 
 
 The top of the chimney should be placed in a position 
 where the products of combustion can pass freely away. 
 If it is placed in an opening or pipe leading to another 
 chimney, care must be taken not to have the draught too 
 powerful, otherwise the heat will be drawn more into the 
 chimney, and the supply of gas in the daytime may be 
 found rather deficient. The furnace will act satisfactorily, 
 though less powerfully, with the chimney standing in an 
 open room without any special outlet for the products of 
 combustion, provided the full height (6 feet) of chimney is 
 employed. Under other circumstances a chimney 4^ or 5 
 feet high may be used. 
 
 This furnace will produce what is generally called a 
 white heat; it will readily melt half a pound of copper, or 
 six ounces of cast iron ; it will melt as large a quantity of 
 those substances as the largest-sized crucible that can be in- 
 troduced into it will contain, sufficient space being reserved 
 around the crucible for draught. It requires from 20 to 30 
 minutes to acquire its highest temperature, and then the 
 entrance part of the chimney exhibits a faint red heat in 
 daylight. If it exhibits much more than this, the draught 
 is too powerful, and, if less, there is not sufficient gas. 
 
 With one dunce of copper put into the cold furnace, 
 and the gas lighted and properly adjusted, the copper 
 generally begins to melt at about the tenth or twelfth 
 minute, and is completely melted by the fifteenth. With 
 the heat well up, 1 ounce of copper has been melted in it 
 in 2J minutes, 1 ounce of cast iron in o minutes, 5 ounces 
 of copper in 4^ minutes, and 3 ounces of cast iron in 5 
 minutes. With the smaller hole in the top of the furnace 
 open, 1 ounce of copper has been melted in 3J minutes, 
 
 H 2 
 
loo GORE'S GAS FURNACE. 
 
 and several ounces of copper have been kept in fusion 
 for upwards of half an hour, and may be kept so for any 
 length of time ; cast iron has also been fused and kept 
 melted under the same conditions. These various effects- 
 have also been obtained in a somewhat diminished degree 
 with the chimney standing in an open room. 
 
 When the small hole D is open some air is drawn in 
 that way, and less air passes up with the gas through the 
 tube 0, but the cold air does not much diminish the 
 temperature of the crucible, because it combines with the 
 excess of gas now passing over the edge of the inner 
 cylinder ; it, however, renders the flame round the crucible 
 white by deficiency of air, and this should be partly cor- 
 rected by lessening the gas. An excess of either gas or 
 air renders the surface of melted copper dull. 
 
 When it is desirable to perfectly avoid contact of air 
 with the fused substance during manipulation, a narrow 
 crucible should be employed, and a thin narrow ring of 
 fire-clay should be placed upon the top of the tube to 
 contract its opening ; the flame then closes completely 
 over the top of the crucible and prevents access of air ; 
 a proper adjustment of gas, together with exclusion of air 
 in this manner, enables a perfectly bright surface of 
 melted copper or even tin to be continuously maintained 
 from which the images of parts above are clearly reflected * 
 The clay ring may be withdrawn by lifting the plate B. 
 A less perfect exclusion of air may be obtained by em^ 
 ploying a narrow crucible placed rather low down in its 
 support. A small iron dish should be placed beneath the 
 tube Z, to receive any melted substance that may fall. 
 The chief conditions of success in the use of this furnace 
 are sufficient gas, a suitable degree of draught, and proper 
 regulation of gas to air. 
 
 Mr. Griffin has devised what he calls a Reverberatory 
 Gas Furnace, which also produces a high temperature 
 without the use of a blowing machine. It is especially 
 suitable for assay purposes on a small scale, and for the 
 decomposition of siliceous minerals by fusion with alkaline 
 carbonates in platinum crucibles, being capable of fusing 
 
REVERBERATORY GAS FURNACE. 101 
 
 1,000 grains of anhydrous sodium carbonate in ten 
 minutes. 
 
 The different parts of this furnace are also arranged in 
 a very convenient manner, so as to admit of its being 
 employed for various purposes in a chemical or assay labo- 
 ratory. It is based upon a new form of gas-burner which, 
 aided by suitable bellows, can be used as a convenient 
 source of heat for most operations of the chemical laboratory 
 and lecture table. It will boil a quantity of liquid exceed- 
 ing two gallons, at once ; it will raise a 4|~-inch fire-clay 
 crucible to full redness ; it will fuse anhydrous sodium car- 
 bonate in greater quantity than is required for the analysis 
 of a siliceous mineral ; and it will melt small quantities 
 of sterling silver. This amount of power is sufficient for 
 most chemical and many metallurgical operations. 
 
 Fig. 47 represents the gas-burner of this apparatus. 
 The gas is supplied by the horizontal tube, whence it 
 passes through a set of small holes into the box a, in 
 which it mixes with atmospheric air that enters freely by 
 the holes shown in the sketch. The gaseous mixture 
 passes up the vertical tube b, and is inflamed at the top, 
 where it burns with a single tall blue flame, which gives 
 no smoke, very little light, but much heat. In this 
 condition the apparatus differs from 'Bunsen's gas-burner' 
 only in size, c represents a thin brass cap, which fits the 
 air-box a, but moves easily round it ; d is a flat cast-iron 
 box with many holes around the margin, and a few small 
 ones on the top. This box fits loosely on the upper part 
 of the tube , and when it is placed on it and the gas is 
 lighted the flame produced consists of a series of radiating 
 jets, forming a horizontal circular flame of about seven 
 inches in diameter. If b gives a single vertical flame. 
 The ring of flame is suited to the purposes of boiling and 
 evaporation ; the single flame, to ignition and fusion. 
 The height of the apparatus represented by fig. 47 is 
 twelve inches ; the bore of the tube b is one inch ; and 
 the diameter of the fire-box d is four inches. 
 
 When a large crucible is to be heated to redness, the 
 gas-burner is to be used without the rose, and is to be 
 
102 
 
 EEVERBERATORY GAS FURNACE. 
 
 FIG. 47. 
 
 arranged with the furnace fittings that are represented in 
 perspective by fig. 48, and in section by fig. 49, and the 
 lower part of fig. 50, a, , c, d. Letter a represents the 
 gas-burner ; fig. 48 b is a tall iron stool ; c a chimney 
 which collects atmospheric air to feed the flame, and leads 
 it up close to the vertical tube of a, by which contrivance 
 the air is warmed and the tube cooled ; d is a furnace-sole 
 or plate of fire-clay ; f is a reverberatory 
 dome, the interior of which is best shown in 
 the section fig. 49 ; c is a cast-iron ring or 
 trivet, represented more clearly in fig. 51 ; 
 g is an iron chimney 24 inches long and 
 3|- inches wide ; and h a damper to lessen 
 the draught when small crucibles are to be 
 heated. The height of this apparatus from 
 
 FIG. 49. 
 
 a to the top of / is 24 inches ; and the external diameter 
 of the dome f is about 8 inches. The crucible, which 
 
REVERBERATORY GAS FURNACE. 103 
 
 may be from 4J to 4j- inches in height, is placed on the 
 iron ring -0, fig. 50, and that on the clay sole d, and it is 
 then covered by the dome /. The gas should be lighted 
 after the crucible is placed in its position and before the 
 dome is put on. The dome and the chimney are then to 
 be added and the operation allowed to proceed. With a 
 crucible of the above size the damper h is not required ; 
 but it must be used when the crucible is under 4 inches in 
 height, otherwise the draught occasioned by the extra 
 space within the dome causes the flame to blow down. 
 The damper must be put on the chimney before the 
 chimney is put on the dome. The iron ring (fig. 51) suits 
 crucibles of different sizes, according to which side, of it is 
 turned uppermost. 
 
 The figures show that a crucible mounted in this 
 FIG. 50. 
 
 FIG. 51. 
 
 furnace can lose very little heat by radiation or conduc- 
 tion, and hence it is that a small gas flame produces a 
 powerful effect. In half an hour a 4f -inch clay crucible, 
 filled and covered, can be heated to full redness. The 
 progress of the ignition can be easily examined by lifting 
 up the chimney g and the dome / by their respective 
 wooden handles. But the action of the furnace can also 
 be judged of by a peculiar roaring noise which it pro- 
 duces. If the gas and air are mixed in due proportions, 
 the roar is regular and continuous ; if there is too much 
 gas the roar is lessened, if too much air the roar is 
 increased, but is rendered irregular and intermittent'. 
 The greater the noise, the greater the heat in the furnace. 
 And when the roar becomes spasmodic the flame is on the 
 
104 
 
 REVERBERATORY GAS FURNACE. 
 
 FIG 52. 
 
 point of blowing down. To prevent that occurrence, the 
 proportion of air must be lessened or that of gas increased. 
 The following arrangement is convenient when small 
 crucibles are to be strongly heated : anhydrous carbonate 
 of soda in quantities exceeding 1,000 grains can be thus 
 readily fused in a platinum crucible, and sterling silver can 
 be melted in a clay crucible. It is also available for igni- 
 tions or fusions in small porcelain crucibles. Fig. 49 re- 
 presents the arrangement of apparatus, as seen in section : 
 a is the gas-burner ; b the stool ; c the air chimney, and d 
 the furnace-sole, as already explained ; i is a cylinder of 
 fire-clay, 4 inches high, and 4^ inches diameter ; k is a 
 fire-clay furnace, in which is placed a small cast-iron ring 
 about 2 inches in diameter, similar in form to that repre- 
 sented by fig. 51, and on this ring the platinum crucible is 
 adjusted ; / is a fire-clay or plumbago reverberatory dome ; 
 and g is the chimney that forms part of the furnace repre- 
 sented by fig. 48. The crucible being adjusted, the gas 
 lighted, and the dome and chimney 
 put on, the lapse of twelve or fifteen 
 minutes, according to the quality and 
 pressure of the gas, suffices for the 
 fusion of 1,000 grains of carbonate of 
 soda in a platinum crucible. At the 
 heat which this furnace produces, the 
 cast-iron ring does not melt nor alloy 
 with the platinum crucible placed 
 upon it. 
 
 By a modification of these ar- 
 rangements, Mr. Griffin has made a 
 gas furnace for melting quantities of 
 lead, zinc, antimony, &c. This is re- 
 presented by fig. 52. The iron crucible 
 will contain nearly 30 Ibs. of lead and 
 about 24 Ibs. of zinc. The burner 
 readily melts these quantities, and then, 
 with a diminished quantity of gas, will 
 keep the metals fluid. The metals being protected from 
 the air suffer little loss by oxidation. Such operations as the 
 
PRINCIPLES OF HEATING BY GAS. 105 
 
 granulation of zinc are performed with this apparatus with 
 great facility; it serves also for baths of fused metal. In a 
 large furnace of this kind, made for a special operation, 
 60 Ibs. of zinc have been melted with ease, and the inventor 
 believes that, used in this manner, the burner is powerful 
 enough to melt a hundredweight of zinc. 
 
 The principles of heating by gas, which have led Mr. 
 Griffin to the construction of these gas furnaces, may be 
 summed up as follows. When a crucible or other solid 
 body is to be heated, it is to be wrapped in a single flame 
 at the point of maximum heat, and loss of heat by radiation 
 and conduction is to be prevented by the interposition 
 of non-conducting materials (plumbago or fire-clay) ; and 
 when liquids are to be boiled or evaporated, particularly 
 when they are contained in vessels of glass or porcelain, 
 the flame is to be broken up into numerous horizontal jets, 
 and these are to be made to supply a large and regular 
 current of highly heated air, by which alone, and not by 
 the direct application of the flame, the vessel that contains 
 the liquid is to be heated. In both cases provision must 
 be made to secure a sufficient draught of air through the 
 furnace, because every cubic foot of gas requires for com- 
 bustion 10 or 12 cubic feet of air, and the gases which 
 have done their duty must be rapidly carried away from 
 the focus of heat. If the steam, the carbonic acid gas, and 
 the free nitrogen which constitute the used-up gases are not 
 promptly expelled, fresh gaseous mixture in the act of 
 producing additional heat by combustion cannot get near 
 the object that is to be heated, and the heat so produced 
 out of place is wasted. 
 
 Bunsen's gas-burner, whatever its size, is subject to 
 two defects : sometimes the flame burns white and smoky, 
 and sometimes it blows down, the gaseous mixture ex- 
 plodes, and the gas then burns with a smoky flame in the 
 tube. The remedies for these defects are as follows : If 
 the flame is white only when the gas is turned on very 
 full, the remedy is to lessen the supply of gas ; but if the 
 flame continues to burn white at the top when the gas is 
 gradually turned off and the mass of flame slowly sinks, 
 
loo BUNSEN'S GAS BURNER. 
 
 then the holes which deliver the gas from the supply pipe 
 into the air-box a (fig. 47) are too large, and are placed 
 too directly under the centre of the vertical tube b (fig. 47), 
 and these defects must be corrected in the instrument. 
 Finally, when the flame blows down, it is because the 
 supply of atmospheric air is too large in proportion to 
 the supply of gas, and their relative proportions must be 
 altered. To effect this alteration the cap c is to be turned 
 round on the air-box a so as partially to close the holes, 
 and thus lessen the supply of air. If, when the gas is 
 alight, the fiaine needs to be lowered, first the supply of 
 air is to be lessened and then the supply of gas. If the 
 flame is to be enlarged, first the supply of gas must be 
 increased and then the supply of air. In short, to prevent 
 the fiame blowing down, the gas must always be placed in 
 excess, and then have the proper quantity of air adjusted 
 to suit it by means of the regulator c. When gas-burners 
 of this description have to be used in a locality where the 
 pressure of the gas is slight, especially in the daytime, 
 there is a constant tendency in the flame to blow down. 
 The best way to prevent that occurrence is to supply the 
 gas by a pretty wide tube, and to see that the current of 
 gas is not checked by a very narrow bore in the plug of 
 an intervening stop-cock, which is often the unsuspected 
 cause of want of pressure in the supply of gas. If this 
 does not suffice to prevent the blowing down of the gas,v 
 the holes which admit the gas from the supply pipe into 
 the box a of the burner should be enlarged, more or less 
 according to necessity. A large supply of gas compen- 
 sates, to some extent, for want of pressure. 
 
 When a steady and long-continued heat is desired from 
 a Bunsen's burner, it is proper to use two stop-cocks and 
 a length of caoutchouc tube between them. One of these 
 stop-cocks is to be affixed to the burner, and the other to 
 the supply pipe. The latter is to be opened wider than is 
 necessary to supply the required quantity of gas, and the 
 former is to be used to regulate the supply to the burner 
 exactly ; under these circumstances, if another stop-cock 
 is opened and gas burnt in the immediate neighbourhood, 
 
METHOD OF MOUNTING CRUCIBLES. 10T 
 
 the flame does not so readily blow down in the regulated 
 burner as it does when only the stop-cock on the supply 
 pipe is used. 
 
 When a crucible is suspended by wires or by a ring 
 over the flame of a spirit lamp or gas-burner, the flame 
 and the hot air supplied by the flame strike the crucible 
 for an instant and then pass away and do no more good. 
 At the same time, the effect of the heating power on the 
 crucible is lessened by other circumstances namely, by 
 radiation on all sides, by a mass of cold air which con- 
 stantly rises around and in contact with it, and by the con- 
 ducting power of the metallic apparatus which supports 
 both the crucible and the lamp. These losses are avoided 
 if the crucible is inclosed in a furnace made of a non- 
 conducting material, such as fire-clay, which can absorb 
 and retain heat. In the description of the gas furnaces, 
 and in that of Mr. Charles Griffin's oil-lamp furnace, several 
 methods of mounting crucibles in fire-clay jackets have 
 been shown ; and we will now describe some of Mr. Griffin's 
 fittings that may be used to construct temporary table fur- 
 naces for crucibles that are to be exposed to the flame 
 produced by gas, oils, or spirit up to a temperature close 
 upon, but not quite up to, a white heat; that is to say, 
 up to a heat that will readily melt anhydrous carbonate of 
 soda and small quantities of silver, and so be fit for many 
 metallurgical operations, but which will not melt copper 
 nor cast iron. 
 
 Fig. 53 represents sections of cylinders of fire-clay 
 which are drawn on a scale of 1 to 8, and have the re- 
 lative heights and bores represented in the figures. The 
 clay pieces that is to say, as many of them as are necessary 
 for a given purpose can be adjusted over a gas flame by 
 means of a tripod (fig. 48) or a clay support. 
 
 The crucible to be operated upon is to be supported on 
 a toothed ring made either of cast iron or fire-clay, such 
 as are represented by figs. 51 and 54. Fig. 51 is a ring 
 of east iron, h representing it in section and i as seen from 
 above. It is about two inches in diameter, and has three 
 teeth projecting towards the middle of the ring. This 
 
^08 
 
 MOUNTS FOR CEUCIBLES. 
 
 FIG. 53. 
 
 ring can be supported by any of the clay cylinders whose 
 bore does not exceed two inches. Fig. 54 is a ring of 
 
 fire-clay of 4 inches ex- 
 ternal diameter, and 1 
 inch in thickness, pro- 
 vided with three teeth 
 that project inwards, 
 and upon which a cru- 
 cible can be supported 
 without injuring the 
 draught of the gas fur- 
 nace. 
 
 Both these grates 
 will support crucibles at 
 the highest temperature 
 which can be produced 
 by spirit, oil, or gas, 
 without a blast of air ; 
 but at a white heat pro- 
 duced by any of these 
 fuels with a blast of air, 
 the iron ring melts, and, 
 if the heat is long con- 
 tinued, those of fire-clay 
 soften and partially give 
 way. When the fire- 
 clay grate (fig. 54) is 
 required to sustain a 
 very high temperature for a considerable time, it is proper 
 to have it made of 6 inches diameter, as represented by 
 fig. 53 j9, the air-way in which is the same 
 as that of the small grate, but the clay ring 
 is much stronger. 
 
 The grate is fixed above the flame at 
 a distance which is found by trial to place 
 the crucible on the point of greatest heat. 
 Commonly a 4-inch cylinder (53, h or g) 
 placed upon a suitable support serves the purpose. The 
 bore of the cylinders at the bottom must be wider than 
 
 FIG. 54. 
 
MOUNTS FOR CRUCIBLES. 109 
 
 the burner, to allow of a considerable influx of atmo- 
 spheric air around the flame. The grate is placed on this 
 cylinder, the crucible on the grate, and then another cylin- 
 der around the crucible. The choice of this upper cylinder 
 depends entirely upon the size of the crucible that is to be 
 heated. Whatever the size of the crucible, the cylinder 
 must be so chosen as to fit the crucible as accurately as 
 possible, leaving between it and the furnace walls an open 
 space of not less than ^ inch, nor more than ^ inch all 
 round. If the upper cylinder is not contracted at the top 
 like 53 efg, then a cylinder of narrow bore, such as 53 a 
 or , must be put upon it, in order to deflect the flame and 
 the rising current of hot air upon the top of the crucible,, 
 and thus produce a reverberatory furnace. Finally, an iron 
 chimney, 2 or 3 feet long, must be put upon the furnace,, 
 to force up a draught of air sufficient to feed the flame. 
 
 Suppose a small rose gas-burner is to be arranged for 
 an ignition, with the use of a fire-clay support, the com- 
 bination of pieces necessary for the purpose may be those 
 represented by figure 55, where A is the fire-clay support, 
 and the rest of the pieces are those which are shown at 
 fig. 53, and described at the letters placed against each of 
 them in this figure. It is evident that the application of 
 this furnace to crucibles of different sizes depends upon 
 the proper choice of the cylinders here marked i and e. 
 Of course there is only a limited choice of crucibles suit- 
 able for such operations. Three inches is the extreme 
 width between the furnace walls of any of the pieces in 
 fig. 53, from a to </, and though larger cylinders^ could be 
 used, such as i to 0, it must be remembered that the flame 
 of a lamp without blast has only a limited power, and that 
 although a given flame will fuse 1,000 grains of carbonate 
 of soda in a platinum crucible, it may only heat to a 
 moderate redness a large clay crucible. Yet, considering 
 that low degrees of heat are suitable for many purposes, it 
 is convenient to have the power of readily adjusting a 
 temporary furnace to the bulk of any crucible which it 
 is desired to heat. 
 
 The clay pieces (fig. 53 i to p) are those that have been 
 
110 
 
 MOUNTS FOE CRUCIBLES. 
 
 expressly designed for the blast oil furnace already 
 described ; but these can also be used for spirit and gas 
 furnaces, the respective sizes being chosen in each case 
 according to the size of the crucible that is to be ignited. 
 
 In respect to the means of supporting a crucible, it has 
 been shown that clay trivets with a wide flange namely, 
 the 6-inch trivets fig. 53 p will support a crucible contain- 
 ing 5 Ibs. of iron until that quantity of iron is melted, even 
 under the operations of a blast ; so that it is evident that 
 this method of supporting a crucible in a gas flame may 
 be always depended upon when no blowing- machine is 
 employed. 
 
 Fig. 56 represents the gas furnace arranged for boiling 
 or evaporating : a is the gas-burner ; b an iron stool with 
 
 FIG. 55. 
 
 FIG. 56. 
 
 three legs; c .a furnace body or iron 
 jacket lined with plumbago or fire-clay. 
 This furnace may be 14 inches high and 
 9 inches in diameter. The three brackets 
 fixed on the upper part of the jacket 
 serve to support the vessel that contains 
 the liquid that is to be boiled or evapo- 
 rated. A porcelain basin of 16 or 18 
 inches in diameter can be thus supported. It is important 
 to allow -between the jacket c and the evaporating basin 
 plenty of space for the escape of the heated air, which 
 ascends from the interior of the furpace. When the 
 evaporating basin is of small diameter, it may be sup- 
 
TEMPORARY GAS FURNACES. Ill 
 
 ported on iron triangles, placed in the furnace c. The 
 section shows that around the vertical tube of the gas- 
 burner a there is in the bottom of the furnace c a circular 
 opening which is of 2 inches diameter, and through which 
 air passes freely, partly to feed the flame and partly to be 
 heated by the flame and be directed upwards in a con- 
 tinuous current upon the lower surface of the basin that 
 is to be heated. The flame within the furnace burns 
 steadily. No side currents of air agitate it. No part of 
 it touches the basin, which should receive its heat solely 
 from the mass of ascending hot air. The gas-burner. thus 
 arranged and supplied by a gas-pipe of J-inch bore, burns 
 about 3 cubic feet of gas in an hour, and the flame which 
 it produces, acting upon water contained in an open porce- 
 lain evaporating basin, will heat from 60 to 212 F. 
 
 1 quart in 5 minutes 
 
 1 gallon in 15 . 
 
 2 gallons in 30 
 
 and when the water boils it is driven off in steam at the 
 rate of more than a gallon of water per hour. The method 
 is consequently applicable to distillation on a small scale, 
 and to numerous other laboratory operations. 
 
 An excellent gas-burner for general laboratory use is 
 Mr. Fletcher's solid-flame burner, shown in fig. 57. This 
 is one of the very best heating burners which has yet been 
 made. The flame is FIG. 57. 
 
 of a brilliant green 
 colour, solid, and of 
 the same tempera- 
 ture throughout ; the 
 usual heating bur- 
 ners having a flame 
 with a hollow centre 
 of unconsumed gas. 
 The pattern shown 
 in the figure mea- 
 sures only 3-^ inches 
 in height, and it will melt half a hundredweight of lead in 
 an iron pot. It will boil half a gallon of water in a 
 
112 SOLID-FLAME BURNER. 
 
 flat copper kettle in five minutes, and will melt 6 Ibs. of 
 lead or solder in an iron ladle in seven minutes. The 
 burner can be adapted for a blast where very high power 
 is required in a small burner. With a blast it can be 
 made to consume any gas supply up to 200 cubic feet per 
 hour, giving a very high duty. One advantage which 
 this burner possesses over those of the ordinary kind with 
 wire-gauze tops is that it cannot be spoiled by any acci- 
 dent. In case a solution boils over, and chokes the holes 
 in the perforated copper dome, the latter can be lifted 
 off (when the burner is warm) and cleaned. The whole 
 burner is designed to stand the roughest and heaviest 
 work without injury. 
 
 For laboratory purposes, where there is always a 
 liability of liquids falling on the burners, Mr. Fletcher has 
 introduced drip-proof high-power burners with pure solid 
 nickel flame surfaces. These are shown at fig. 58. 
 
 These are undamaged by the dirtiest work, and will 
 burn perfectly under a constant drip. The nickel flame 
 
 FIG. 58. 
 
 DRIP PROOF 
 
 HIGH POWER BURNER 
 
 surface adds considerably to the first cost of the burner, 
 but it is practically everlasting, and will neither rust nor 
 burn away. 
 
 These burners are generally used under vessels either 
 fixed or supported on wrought-iron stands. The burners 
 themselves are very small in proportion to the power 
 and the size of vessel they will heat. The bottom of 
 the vessel should be about 1-J inch clear above the top 
 of the burner. 
 
LUTES AND CEMENTS. 113 
 
 An improved pattern of Safety Bunsen Burner, also 
 introduced by Mr. Fletcher, is shown in fig. 59. 
 
 This will be found FIG 59 
 
 as perfect as any up- 
 right tube burner can 
 possibly be made, of 
 the highest possible 
 power for size, can be 
 turned down to the 
 merest flicker without 
 lighting back, and can 
 be mounted on tubes 
 in any form or number 
 when very high powers 
 are required. 
 
 The india-rubber 
 tubing for these burners 
 must be of good size, 
 and smooth inside, made 
 without wire. 
 
 LUTES AND CEMENTS. 
 
 It may be as well 
 to mention in this part 
 of the work the various 
 lutes and cements which 
 may be employed, either in fire operations or in making 
 good joints in experiments with gases or liquids. The 
 following are the principal kinds. 
 
 FIRE LUTE is composed of good clay two parts, sharp 
 washed sand eight parts, horse-dung one part. These 
 materials are to be intimately mixed ; and afterwards the 
 whole is to be thoroughly tempered like mortar. Mr. 
 Watt's fire lute is an excellent one, but is more expensive. 
 It is made of fine powdered Cornish (porcelain) clay 
 mixed to the consistence of thick paste, with a solution of 
 borax in the proportion of 2 ounces of borax to a pint of 
 hot water, 
 
 FAT LUTE is prepared by mixing fine clay, in a fine 
 
 I 
 
114 LUTES AND CEMENTS. 
 
 powder, with drying oil, so that the mixture may form a, 
 ductile paste. When this paste is used, the part to which 
 it is applied ought to be very clean and dry, otherwise it 
 will not adhere. Glazier's putty is very similar to this. . 
 
 ROMAN CEMENT. This must be kept in well-closed 
 vessels, and not moistened until it is required for use. 
 
 PLASTER OF PARIS. This is mixed with water, milk, or 
 weak glue, or starch water. 
 
 These three lutes stand a dull red heat : the two latter 
 may be rendered perfectly impermeable to gaseous bodies 
 by being smeared over with oil, or a mixture of oil and 
 wax. 
 
 LINSEED OR ALMOND MEAL, mixed to the consistence of a 
 paste with water, milk, lime water, or starch paste. This 
 lute is very manageable and impermeable, but does not 
 withstand a heat greater than about 500 P. 
 
 LIME AND EGG LUTE If just the sufficient quantity of 
 
 water be added to quick lime to reduce it to a dry powder, 
 and that is mixed well and rapidly with white of egg 
 .diluted with its own volume of water, and the mixture 
 spread immediately on strips of linen and applied to the 
 part, then powdered with quick lime, it forms a good 
 cement. Instead of white of egg, lime and cheese may 
 be used, or lime with weak glue water. This lute dries 
 very rapidly, becoming very hard and adhering strongly 
 to glass ; but its great inconvenience is the want of 
 flexibility. 
 
 WHITE LEAD MIXED WITH OIL. If this mixture be spread 
 upon strips of linen, or bundles of tow, it acts much in the 
 same manner as the lime lutes. 
 
 YELLOW WAX is often used as a lute, but it becomes 
 very brittle at a low temperature. It may be rendered 
 less brittle, and at the same time more fusible, by an 
 admixture of one eighth crude turpentine. 
 
 SOFT CEMENT is prepared by fusing yellow wax with 
 half its weight of crude turpentine and a little Venetian 
 red in order to colour it. It is very flexible, and takes 
 any desired form under the pressure of the fingers. 
 
 CEMENT FOR BRASS ON GLASS (for instance, petroleum 
 
LUTES AND CEMENTS, 115 
 
 lamps). 1 part of caustic soda, 3 parts of resin, arid 5 
 parts of water are boiled together till solution is effected, 
 when it is. intimately mixed with one-half of its weight in 
 plaster of Paris. The mixture hardens within one hour, 
 and is impermeable for coal oil. 
 
 CEMENT FOR MENDING PESTLES &c. One of the strongest 
 cements and one that can be very readily made is ob- 
 tained when equal quantities of gutta-percha and shellac 
 are melted together and well stirred. This is best done in 
 an iron capsule placed on a sand-bath, and heated either 
 over a gas furnace or on the top of a stove. It is a com- 
 bination possessing both hardness and toughness, qualities 
 that make it particularly desirable in mending pestles and 
 mortars. It is very useful for securing the handles to the 
 wedgwood ware, and some old ones that were much 
 chipped and split, when thus mended, have been quite as 
 useful as new ones, and have stood several months' wear 
 without any apparent change. When this cement is used 
 the articles to be mended should be warmed to about the 
 melting point of the mixture and then retained in proper 
 position until cool, when they are ready for use. 
 
 ADHESIVE PASTES. 1. Tragacanth, 1 oz. ; gum arabic, 
 4 ozs. ; water, 1 pint. Dissolve, strain, and add thymol, 14 
 grains ; glycerin, 14 ozs. ; and water to make 2 pints. 
 Shake or stir before using it. 
 
 2. Eye-flour, 4 ozs. ; alum, -J oz. ; water, 8 ozs. Eub to 
 a smooth paste, pour into a pint of boiling water, heat until 
 thick, and finally add glycerin, 1 oz. and oil of cloves, 30 
 drops. 
 
 3. Eye-flour, 4 ozs. ; water, 1 pint. Mix, strain, add 
 nitric acid, 1 drachm ; heat until thickened, and finally add 
 carbolic acid 10 minims, oil of cloves, 10 minims, and 
 glycerin, 1 oz. 
 
 4. Dextrin, 8 parts ; water, lOparts ; acetic acid, 2 parts. 
 Mix to a smooth paste and add alcohol, 2 parts. This is 
 suitable for bottles or wood, but not for tin, for which, 
 however, the first three are adapted. 
 
 5. The paste used by the United States Government 
 for gumming postage stamps is made by the following 
 
 i 2 - 
 
116 LUTES AND CEMENTS. 
 
 formula. It has the properties of being very adhesive, does 
 not become brittle or scale off, and is well adapted for 
 sticking paper labels to tin and other metals. Take of 
 starch, 2 drachms ; white sugar, 1 ounce ; gum arabic, 2 
 drachms ; water, q.s. Dissolve the gum, add the sugar, 
 and boil until the starch is cooked. 
 
 WATERPROOF CEMENT. Mr. Edmund Davy, F.K.S., de- 
 scribed a cement made by melting in a saucepan two 
 parts by weight of common pitch, and adding to it one 
 part by weight of gutta-percha, stirring and mixing them 
 well together until they were completely incorporated 
 with or united with each other. The mixture then forms 
 a homogeneous fluid which may be used in this state for 
 many purposes, and is remarkable on account of the 
 facility and tenacity with which it adheres to metals, 
 stones, and glass. It may be poured into a large basin of 
 cold water, in a thinner or thicker stream, or as a cake. 
 In this state, while warm, it is quite soft, but may be soon 
 taken up out of the water and drawn out into longer or 
 pressed into shorter pieces, or cut or twisted into frag- 
 ments, which may again be readily reunited by pressure. 
 When the cement is cold, or before, it may be removed 
 from the water and wiped dry, when it is fit for use. It 
 is of a black colour ; when cold, it is hard. It is not 
 brittle, but has some degree of elasticity, which is in- 
 creased by a slight increase of heat. It appears to be not 
 so tough as gutta-percha, but more elastic. Its tenacity 
 is very considerable, but inferior to gutta-percha. It 
 softens when put into water at about 100 F. ; when 
 heated to above 100 F. it becomes a thin fluid ; and if 
 the heat is gradually increased, it passes through inter- 
 mediate stages of softness, becomes viscous like bird-lime, 
 and may be extended into threads of indefinite length : it 
 remains in this state even when exposed for some time in 
 a crucible to the heat, of boiling water, 202 F. Water 
 appears to have no other action upon it but that of soften- 
 ing it when warm or hot, and slowly hardening it when 
 cold* The cement adheres strongly, if pressed on metal 
 or other surfaces, though water be present, provided such 
 
LUTES AND CEMENTS. 117 
 
 surfaces be warm. This cement is applicable to many 
 useful purposes. It adheres with great tenacity to metals, 
 wood, stones, glass, porcelain, ivory, leather, parchment, 
 paper, hair, feathers, silk, woollen, cotton, linen fabrics, 
 &c. It is well adapted for glazing windows, or as a 
 cement for aquariums. This cement does not appear to 
 affect water, and it will apparently be found applicable 
 for .coating metal tanks ; to secure the joints of stone 
 tanks ; to make a glue for joining wood, which will not 
 be affected by damp ; and to prevent the depredations of 
 insects on wood. 
 
 RESINOUS OR HARD CEMENT is made by fusing together 
 at the lowest possible temperature one part of yellow wax 
 and five or six of resin, and then adding gradually one part 
 of red ochre or finely powdered brickdust (plaster of Paris 
 succeeds very well), and then raising the temperature to 
 212 at least, until no more froth arises or agitation takes 
 place, and stirring it continually until cold. This cement 
 is employed in a hot state. This lute is much used for 
 fixing brass caps, &c., to air-jars. 
 
 CAOUTCHOUC. Tubes of vulcanised caoutchouc form a 
 very ready means of attaching one piece of apparatus to 
 another, and they possess the peculiar advantage of flexi- 
 bility, which allows the various parts of the apparatus 
 which they connect to move in different directions to 
 a slight extent, so that the whole is not so likely to be 
 fractured as when connected in an inflexible manner. 
 Caoutchouc is also less acted upon by gases and vapours 
 than almost any other substance w^e know ; even chlorine 
 attacks it but slowly, and when unvulcanised it possesses 
 the valuable property of forming a perfect joint when 
 freshly cut edges are brought and pressed together, hence 
 the facility with which it is manufactured into tubes. The 
 
 mode of manufacturing small connecting tubes, which are 
 
 < . 
 
 often required to be of unvulcanised caoutchouc, is as 
 follows : Take a piece of the sheet caoutchouc of the 
 required size, and warm it either in the hand or before a 
 fire, until it is perfectly soft ; then place it around a glass 
 rod of the requisite size, pressing the edges close together 
 
118 LUTES AND CEMENTS. 
 
 with the fingers ; when close together cut off the super- 
 abundance with a sharp pair of scissors, and the newly 
 cut edges will unite by simple pressure of the nail. When 
 well executed the joint is scarcely apparent. In order 
 to prevent the caoutchouc from adhering to the rods on 
 which the tube is formed, a little moisture or dry starch 
 may be employed. When caoutchouc is not at hand, 
 oiled paper may be substituted, the joint being made of 
 wax. 
 
 Faraday gives the following directions for luting iron, 
 glass, or earthenware retorts, tubes, &c. ; for furnace opera- 
 tions. When the lute has to withstand a very high tem- 
 perature it should be made of the best Stourbridge clay, 
 which is to be made into a paste varying in thickness 
 according to the opinion of the operator. The paste 
 should be beaten until it is perfectly ductile and uniform, 
 and a portion should then be flattened out into a cake of 
 the required thickness, and of such a size as shall be most 
 manageable with the vessel to be coated. If the vessel be 
 a retort or flask, it should be placed in the middle of the 
 cake, and the edges of the latter raised on all sides and 
 gradually moulded and applied to the glass ; if it be a 
 tube, it should be laid on one edge of the plate, and then 
 applied by rolling the tube forward. In all cases the 
 surface to be coated should be rubbed over with a piece 
 of the lute dipped in water for the purpose of slightly 
 moistening and leaving a little of the earth upon it; if 
 any part of the surface becomes dry before the lute is 
 applied, it should be re-moistened. The lute should be 
 pressed and rubbed down upon the glass successively 
 from the part where contact was first made to the edges, 
 for which purpose it is better to make them thin by 
 pressure and also somewhat irregular in form, and if 
 at all dry they should be moistened with a little soft 
 lute. The general thickness may be about J to ^ oi an 
 inch. 
 
 Being thus luted, the vessels are afterwards to be 
 placed in a warm situation, over the sand-bath or near 
 the ash-pit, or in the sun's rays. They should not be 
 
LUTES AND CEMENTS. 119 
 
 allowed to dry rapidly or irregularly, and should be 
 moved now and then to change their positions. To pre- 
 vent cracking during desiccation, and the consequent 
 separation of the coat from the vessel, some chemists 
 recommend the introduction of fibrous substances into 
 the lute, so as mechanically to increase the tenacity of its 
 parts. Horse-dung, chopped hay and straw, horse- and 
 cow-hair, and tow cut short, are amongst the number. 
 When these are used, they should be added in small 
 quantity, and it is generally necessary to add more 
 water than with simple lute, and employ more labour 
 to insure a uniform mixture. It is best to mix the 
 chopped material with the clay before the water is put- 
 to it, and, when adding the latter, to mix at first by 
 stirring up the mass lightly with a pointed stick or fork ; 
 it will then be found easy, by a little management, to 
 obtain a good mixture without making it very moist. 
 
 The luting ought to be made as dry as possible con- 
 sistent with facility in working it. The wetter it is, the 
 more liable to crack in drying, and vice versa. 
 
 Mr. Willis recommends, when earthenware retorts, 
 &c., are to be rendered impervious to air, the following 
 (-oating : One ounce of borax is to be dissolved in half a 
 pint of boiling water, and as much slaked lime added as 
 will make a thin paste. This composition is to be spread 
 over the vessel with a brush, and, when dry, a coating of 
 slaked lime and linseed oil is to be applied, This will dry 
 sufficiently in a day or two, and is then fit for use. 
 
 IRON CEMENT. This mixture is used for making per- 
 manent joints generally between surfaces of iron. Clean 
 iron borings or turnings are to be slightly pounded, so as 
 to be broken but not pulverised : the result is to be sifted 
 coarsely, mixed with powdered sal-ammoniac and sulphur, 
 and enough water to moisten the whole slightly. The 
 proportions are 1 sulphur, 2 sal-ammoniac, and 80 iron. 
 No more should be mixed than can be used at one time. 
 Mr. Cooley states that he is informed by one of the first 
 engineers in London that the strongest cement is made 
 without sulphur, and with only one or two parts of sal- 
 
120 PLUMBAGO AND CLAY CEUCIBLES. 
 
 ammoniac to 100 of iron borings ; but that when the 
 work is required to dry rapidly, as for the steam joints of 
 machinery wanted in haste, the quantity of sal-ammoniac 
 is increased a little, and a very small quantity of sulphur 
 is added. This addition makes it set quicker, but reduces 
 its strength. 
 
 Several excellent cements are described in Cooley's 
 ' Cyclopedia of Practical Eeceipts.' From these the fol- 
 lowing are selected : 
 
 BEALE'S CEMENT. Chalk, 60 parts ; lime and salt, ol 
 each 20 parts ; Barnsey sand, 10 parts ; iron filings or 
 dust, and blue or red clay, of each 5 parts. Grind together 
 and calcine. This is patented as a fire-proof cement. 
 
 BOILER CEMENT. Dried clay in powder, 6 Ibs. ; iron 
 filings, 1 lb., made into a paste with boiled linseed oil. 
 This is used to stop leaks and cracks in iron boilers, 
 stoves, &c. 
 
 BRUYERE'S CEMENT. Clay, 3 parts ; slaked lime, 1 part ; 
 mix and expose them to a full red heat for 3 hours, then 
 grind to powder. 
 
 This makes a good hydraulic cement. 
 
 OXYCHLORIDE OP ZINC CEMENT. In solution of zinc chlo- 
 ride, of 1-49 to 1-65 specific gravity, dissolve 3 per cent, 
 of borax or sal-ammoniac, and then add zinc oxide, which 
 has been heated to redness, until the mass is of a proper 
 consistency. This cement becomes as hard as marble. It 
 may be cast in moulds like plaster of Paris. 
 
 CRUCIBLES, CUPELS, ETC. 
 
 The crucibles best known in commerce are the Hessian, 
 the Cornish, the Stourbridge, and the London clay cru- 
 cibles ; charcoal, porcelain, plumbago, platinum, iron, 
 nickel, silver, and gold crucibles are also required in small 
 operations. Of the clay crucibles, the London pots are 
 much to be preferred, on account of their very refractory 
 nature. They resist the action of fused oxide of lead 
 better than most clay crucibles, and they are also 
 better made than the two other kinds, being much 
 
PLUMBAGO AND CLAY CRUCIBLES. 121 
 
 smoother and more regularly formed. They have the 
 form of a triangular pyramid (see fig. 60, crucibles and 
 cover), and are made in such sizes that they fit one into 
 the other, forming nests. The triangular form is very 
 convenient, because there are three spouts, from either of 
 which can be poured the fused contents of the pot. The 
 Cornish crucibles are circular, and do not stand changes 
 of temperature so well as the London pots, neither can 
 they endure such an extreme FIG. 60. 
 
 heat, for they agglutinate 
 and run together at a tempe- 
 rature which does not touch 
 the others. Dr. Percy says 
 they are more generally useful 
 than any other crucible. The 
 Hessian pots are the worst of 
 all ; they do not stand mode- 
 rate change of temperature 
 without risk of fracture, so 
 that they require to be very 
 carefully used. There is also another kind of pot in use, 
 made of the same material as the London crucibles, termed 
 a ' skittle pot,' from its resemblance to the ordinary wooden 
 skittle or ninepin. They are exceedingly useful for the 
 fusion of large masses of matter, or such substances as 
 boil or bubble much when heated. Plumbago or graphite 
 crucibles are rapidly superseding all other kinds when 
 metals have to be melted. They possess many advan- 
 tages over clay crucibles. Their surface is very smooth ; 
 they are not liable to crack, however violent the changes 
 of temperature may be to which they are subjected ; they 
 bear the highest heat without softening, and can be used 
 repeatedly. Owing to the reducing property of the carbon 
 they contain, they must not be employed when oxidising 
 actions are required. 
 
 Crucibles and all plumbago fittings for furnaces should 
 be of the ' Salamander ' brand. These require no anneal- 
 ing or care in heating up, and stand strong fluxes better 
 than the ordinary make. 
 
2-2 PLUMBAGO AND CLAY CRUCIBLES. 
 
 The Patent Plumbago Crucible Company have re- 
 cently introduced a very excellent fluxing crucible. It is 
 made of fine white china clay, is perfectly smooth inside 
 and out, and will stand very high temperatures without 
 softening. 
 
 Stou'rbridge clay crucibles are not much used. They 
 require the greatest care in using them, and are spoilt 
 after the first operation. 
 
 Porcelain crucibles are not used in large assaying or 
 metallurgical operations, but they are invaluable in small 
 laboratory experiments. They are practically infusible, 
 are little liable to crack, and are almost unacted on by 
 reagents and fluxes. In many cases they will replace the 
 more expensive platinum crucibles, and where easily re- 
 ducible metals are under treatment they must be used in 
 preference to platinum. 
 
 Crucibles, in order to be perfect and capable of being 
 used indifferently for any operation, ought to possess the 
 four following qualities : first, not to break or split when 
 exposed to sudden changes of temperature ; secondly, to 
 be infusible ; thirdly, to be only slightly attacked by the 
 fused substances they may contain ; fourthly and lastly, to 
 be impermeable, or nearly so, to liquids and gases. But 
 as it is very difficult to unite all these qualifications, 
 various kinds of pots are made to fulfil one or more of 
 them. 
 
 In order to render crucibles capable of withstanding 
 changes of temperature without breaking, a certain pro- 
 portion of substances infusible by themselves is mixed 
 with the pasty clay; sand, flint, fragments of old crucibles, 
 black-lead, and coke are used for this purpose. They are 
 reduced to a state of division more or less fine, according 
 to the grain of the clay paste. For ordinary pots the 
 powder ought not to be very fine ; but for porcelain cru- 
 cibles it ought to be as fine as flour. The choice of these 
 various bodies depends upon the use for which the crucible 
 is intended. 
 
 The most refractory crucibles are those made with the 
 pure clays, or such as contain little or no oxide of iron r 
 
PLUMBAGO AND CLAY CRUCIBLES. 123 
 
 and especially those free from calcareous matters. Amongst 
 clays, the best are those which contain most silica ; never- 
 theless, these are not absolutely infusible, and in the high 
 temperature of a wind furnace they sometimes soften so 
 much as actually to fall into a shapeless mass. This defect, 
 as before stated, can be in some measure diminished by 
 mixing with the clay a quantity of graphite or coke ; 
 either of these substances forms a kind of solid skeleton, 
 which retains the softened clay, and prevents its falling 
 out of shape. 
 
 Coke and black-lead are more efficacious than sand, 
 because they have no action on clay, whilst sand forms a 
 fusible compound with it. If too large a quantity of 
 black-lead or coke be employed, it gradually consumes in 
 the fire, and the pots become porous, and break at the 
 least movement. Wood charcoal can be used instead of 
 black-lead or coke, but is not so good, as it burns more 
 readily. 
 
 Black-lead crucibles are generally composed of 1 
 part of refractory clay, and from 2 to 3 of black-lead. 
 These pots withstand all possible changes of tempera- 
 ture without cracking, and their form is rarely changed 
 by the heat ; not because they are absolutely infusible, 
 but because they are supported by the skeleton of 
 graphite. 
 
 Crucibles into whose composition carbonaceous matters 
 enter, reduce any oxides that may be heated in them, and 
 hence are inconvenient in certain cases. They can, never- 
 theless, be employed in all cases by giving them a lining 
 of clay, which must be tolerably thick, and well dried 
 before use. 
 
 Earthen crucibles which have not been baked at a 
 white heat are more or less permeable to liquids and 
 gases, according to the grain. In order to render them 
 impermeable to liquids, they must be heated to such a 
 temperature as will suffice to fuse the outside. When 
 treated in this way, however, they are very liable to crack 
 with sudden changes of temperature : the best method, 
 therefore, of rendering them capable of containing water, 
 
124 GOOD AND BAD CRUCIBLES. 
 
 <fcc., is to coat them with the mixture of borax and lime 
 described as Willis's lute. 
 
 In order that crucibles may resist the corrosive action 
 of the fused substances contained within them, they must 
 be as compact as possible, and the substance of which they 
 are made must have little or no tendency to combine with 
 the fused contents. The metals and their non-oxidised 
 compounds attack neither clay nor black-lead ; but there 
 are, nevertheless, some metallic substances galena, for 
 instance which, without exercising any chemical action 
 on earthy matters, have the property of filtering through 
 their pores. 
 
 The readily reducible oxides gradually corrode black- 
 lead crucibles and those pots into the composition of which 
 coke enters, by burning the carbonaceous matter. The 
 greater number of these oxides, the alkalies, earths, and 
 glasses (which are fusible silicates, borates, &c.), act more 
 or less powerfully on the earthy base of all crucibles ; so 
 that these substances are most difficult to keep in fusion 
 for any length of time. They attack the crucible layer 
 by layer, dissolving the substance of which it is composed, 
 and after a lapse of time rendering it so thin that it can- 
 not withstand the pressure of the molten mass within it ; 
 and the fracture of the pot, and consequent loss of con- 
 tents, is inevitable. 
 
 Under tlie same circumstances, all those crucibles 
 whose texture is loose are more readily corroded than 
 those with a firm, compact body ; because the corrosive 
 substance filters to a certain depth in the former crucibles, 
 and in conseqiience has a larger surface to act upon than 
 when it is contained in a compact pot. 
 
 Earthen crucibles may be assayed by noticing the time 
 they will contain fused litharge, which exercises a very 
 corrosive action on them, honeycombing them in all di- 
 rections ; and those pots which contain it longest without 
 undergoing much damage, may be considered the best. 
 However, this method of assay is not exact, even by taking 
 into account the thickness of the pot, for litharge runs 
 through crucibles : first, because it is very fusible, and 
 
GOOD AND BAD CRUCIBLES. 125 
 
 easily filters through their pores ; and secondly, it has the 
 property of forming fusible compounds with all the sili- 
 cates by combining with them. From these remarks it 
 will be evident that a crucible whose grain is loose will 
 readily allow litharge to pass through it, however slightly 
 its substance may be fusible or acted on ; or, on the con- 
 trary, it may be very easily acted on (even when infusible) 
 when it has an extremely fine grain ; so that the prompti- 
 tude with which a crucible is pierced by litharge bears 
 no relation to its fusibility. A crucible of pure quartz 
 will be very readily attacked by litharge, because the 
 latter has much affinity for silica, and the simple silicates 
 of lead are all very fusible ; whilst a crucible composed of 
 silica, alumina, and lime, which by itself is very fusible, 
 would be corroded less rapidly, because the oxide of lead 
 has much less affinity for the earths than it has for the silica; 
 moreover, it forms less fusible compounds with the earths 
 than with silica alone. The assay of crucibles with li- 
 tharge, if not of use in ascertaining their degree of fusi- 
 bility, fulfils perfectly its object when it is wished to prove 
 the resistance a crucible has to the corrosive action of 
 various bodies in a state of fusion ; for of all fusible sub- 
 stances, none exercises such a powerful action on earthy 
 matters as litharge. 
 
 Crucibles ought not only to resist the corrosive action 
 of those bodies they may contain, but also that of the ash 
 produced by the combustion of the fuel in which they 
 may be placed. These ashes being often calcareous, alka- 
 line, or ferruginous, act on the clayey part of the crucibles 
 exactly as the fluxes. Whence it follows, that those cru- 
 cibles which contain litharge longest will also resist the 
 action of the fluxes best. 
 
 In order to ascertain the fusibility of a crucible, a 
 direct experiment must be made, either by heating a piece 
 in a crucible lined with charcoal, and ascertaining if its 
 angles be rounded, if its substance has become translucent, 
 &c. ; or, better still, by heating the crucible to be assayed 
 with another whose properties are well known. 
 
 As to permeability, it may approximately be ascer- 
 
126 CHARCOAL CRUCIBLES. 
 
 tained by filling two crucibles with water, and noting what 
 length of time is required to empty them completely ; the 
 crucible which contains it longest being, of course, the 
 least permeable. 
 
 To prove if a crucible be able to sustain great changes 
 of temperature without breaking, introduce it, perfectly 
 <>old, into a furnace full of lighted coal : take it out when 
 reddish white, and expose it to a current of cold air pro- 
 duced by a bellows or otherwise : if it stand these trials, 
 it may be heated afresh and plunged red-hot into water, 
 and, if it be not broken, placed immediately in the fire. 
 The best pots support all these operations without break- 
 ing ; but it often happens that they are filled with innu- 
 merable small fissures, through which fused matters can 
 pass. This can be ascertained by fusing rapidly in the 
 assay pot a quantity of litharge : if these fissures be pre- 
 sent, the fused oxide will readily filter through them. 
 
 CHARCOAL CRUCIBLES. As all oxidised matters act readily 
 on clay pots, and a great number of the metals and their 
 compounds adhere to them, they have long since been 
 replaced, under certain circumstances, by charcoal cru- 
 cibles, which do not possess these disadvantages. The 
 older assayers used merely a piece of charcoal, with a 
 hole made in it, and then bound round with iron or other 
 wire. The use of these has, however, been abandoned for 
 some time, and earthenware crucibles lined with charcoal 
 have been substituted (see fig. 61, a, 6, and c). These may 
 be considered as charcoal pots enveloped with refractory 
 clay ; they are solid, always free from cracks, and easy of 
 preparation, and they have the same properties as the 
 solid charcoal crucibles without their inconveniences. 
 
 In order to prepare these crucibles, the charcoal must be 
 chosen carefully, so as to contain no foreign substances ; it 
 must be pulverised and passed through a sieve ; the powder 
 moistened with water or treacle, mixed with a spatula, and 
 then kneaded with the fingers until it just adheres, and 
 forms adhesive lumps without being sufficiently wet to 
 adhere to the hand. Some advise the addition of gum to 
 the water with which the charcoal is moistened. The 
 
LINED CRUCIBLES. 127 
 
 crucible is moistened slightly by being plunged into water, 
 and withdrawn as speedily as possible, and about half an 
 inch in depth of the charcoal paste, prepared as above, is 
 placed in it ; the paste is then pressed firmly down, by 
 means of a wooden pestle : the blows are to be slight at 
 first, and then increase in force until it is as firm as possible: 
 another layer is then applied and pressed as before, and 
 the process repeated until the crucible is quite full, taking 
 great care to render all as firm as possible, especially at 
 the sides. In order to make each layer adhere firmly to 
 the other, the surface must be scratched rather deeply 
 
 with the point of a knife before a new layer is applied. 
 When the crucible is completely filled, a hole is to be 
 scooped in the charcoal of about the form of the pot. The 
 sides are then rendered smooth by friction w r ith a glass 
 rod. This is absolutely necessary, so that the metallic 
 globules produced in an assay may not be retained by the 
 asperities of the lining, but may be readily enabled to unite 
 into one button. When a lined charcoal pot is well made, 
 its sides are very smooth and shining. For ordinary use, 
 the lining may be f ths of an inch thick at the bottom and 
 ^th or so at the sides : but in some cases, for instance, 
 when the substance to be fused is capable of filtering 
 through the lining and attacking the pot as a flux, it must 
 be at least twice the above thickness in every part. 
 
 Lined crucibles have many advantages over plain 
 
128 LIME CRUCIBLES. 
 
 crucibles. The lining gives them greater solidity, and 
 prevents a loss of shape when softened ; for plain crucibles 
 are always three-fourths empty when their contents are 
 fused, on account of contraction in volume : the pots 
 then have nothing to sustain their sides when they soften 
 towards the end of the assay, at which period the highest 
 temperature is employed. Besides, vitreous matters do 
 not penetrate the lining, and, exercising no action on it, 
 can be obtained in a state of purity, and the exact weight 
 determined : if fused in a plain pot, the weight could not 
 be ascertained, because a portion would adhere to the 
 sides, and the resulting mass would not be pure, having 
 taken up a portion of the crucible in which the fusion was 
 effected. 
 
 The lining, too, effects the reduction of certain metallic 
 oxides by cementation, and does away with the necessity 
 of adding powdered charcoal to the body to be reduced. 
 This property is very valuable, because, when an oxide is 
 reduced by mixing it with charcoal, an excess must always 
 be employed, and this excess remains with the metal, and 
 prevents its exact weight from being ascertained. No 
 oxidising substances or bodies which readily part with 
 oxygen (oxide of copper, for examp]e) must be calcined in 
 a plumbago or charcoal-lined crucible, unless indeed the 
 chemical union of the charcoal with the oxygen is desired. 
 
 LIME CRUCIBLES. Some years ago Deville proposed the 
 use of crucibles cut out of solid blocks of pure lime, in 
 order to prevent the introduction of carbon and silicon 
 into metals and alloys during the process of fusion. 
 
 The results of experiments made with such crucibles 
 were found to be extremely satisfactory, and metals fused 
 therein, as iron, manganese, nickel, cobalt, &c., were 
 obtained far purer and more malleable and ductile than 
 when fused in the usual clay or brasqued crucibles. Un- 
 less, however, the crucibles required were of very small 
 size, it was found difficult to obtain blocks of lime for 
 shaping them sufficiently large and free from flaws ; and 
 experiment showed a considerable loss, both by breakage 
 when shaping them, and by their cracking when in the 
 
\v 
 
 TY }} 
 
 IJME CRUCIBLES. 129 
 
 furnace. In order to obviate this, trials were made with 
 clay crucibles lined with lime, but ineffectually, as these 
 crucibles invariably melted down before the requisite heat 
 was arrived at a result due to the action of the lime 
 itself upon the outer clay crucible. 
 
 Mr. David Forbes, F.R.S.,has published in the c Chemi- 
 cal News ' the result of some very valuable experiments 
 on this subject. The arrangement he proposes fully 
 answers the purpose, the crucibles being capable of stand- 
 ing the heat of melted wrought iron or cobalt without 
 fusion or cracking, as well as of being made of any reason- 
 able size. 
 
 A clay crucible of somewhat larger capacity than the 
 desired lime one is filled with common lamp-black, com- 
 pressing the same by stamping it well down. The centre 
 is then cut out with a knife until a shell or lining of lamp- 
 black is left firmly adherent to the sides of the crucible, 
 and about half an inch or less in thickness, according to 
 the size of the crucibles ; this lining is now well rubbed 
 down with a thick glass rod until its surface takes a fine 
 glaze or polish, and the whole cavity is then filled up with 
 finely powdered caustic lime, thoroughly pressed down, 
 and a central cavity cut out as before ; or the lime powder 
 may be at once rammed down round a central core of the 
 dimensions of the intended lime crucible. 
 
 This lime lining is naturally rather soft before being 
 placed in the furnace, but upon heating, it agglutinates, and 
 forms a strong and compact crucible, which is prevented 
 from acting upon the outer one by the interposed thin layer 
 of lamp-black, and at the end of the experiment it generally 
 turns out as solid and compact as those made in the lathe. 
 
 Experiments made with such crucibles, even up to 
 dimensions containing several pounds of metal, have proved 
 them extremely well suited for these operations, and doubt- 
 less similar crucibles could be made, lined with magnesia 
 or alumina as required. In some cases ordinary black- 
 lead crucibles, lined with powdered lime, magnesia, or 
 alumina, might possibly be found to answer. 
 
 Having frequently used lime crucibles in metallurgical 
 
 K 
 
130 ALUMINA CRUCIBLES. 
 
 operations, and having met with the inconvenience pointed 
 out by Mr. Forbes, the editor can appreciate the great 
 value of his improvement. It is one which cannot fail to 
 be extensively adopted in metallurgical laboratories. 
 
 In certain particular experiments, crucibles are lined 
 with other bodies besides charcoal and lime, such as silica, 
 alumina, magnesia, or chalk, by merely moistening their 
 respective powders with water, and applying the paste as 
 above described for the charcoal. A slight layer of chalk 
 lessens the liability of attack from fused litharge. 
 
 ALUMINA CRUCIBLES are strongly to be recommended in 
 many metallurgical operations. They are made in the 
 following manner. Ammonia alum is ignited at a full 
 white heat, when it leaves behind pure alumina in a 
 dense compact form : this is to be finely powdered. To a 
 solution of another portion of ammonia alum in water, 
 ammonia is added, when alumina is precipitated in a gela- 
 tinous state : this is to be washed until free from sulphate 
 of ammonia. The dense alumina is then mixed with water 
 and worked up into a paste, the precipitated gelatinous 
 alumina being kneaded in from time to time ;' this gives 
 coherency : and when sufficient has been added (which 
 must be ascertained experimentally), the mass may be 
 moulded into shape. These crucibles require slow and 
 careful drying ; but they well repay all the care which is 
 bestowed on them, for they do not readily crack, are 
 attacked by very few fluxes, give out no impurities to 
 metals which are melted in them, and are infusible at the 
 highest heat of the furnace. 
 
 MAGNESIA CRUCIBLES AND BRICKS. M. H. Caron has 
 pointed out the advantages which would accrue to metal- 
 lurgy from the employment of magnesia as a refractory 
 material. Formerly the high price of this earth appeared 
 likely to confine it to the laboratory. Now, circumstances 
 have happily changed ; recent modifications introduced 
 into the manufacture of cast steel, and especially the 
 employment of Siemens's furnace and Martin's process, 
 absolutely demand more refractory bricks than those at 
 present in use, irrespective of price. On the other hand, 
 
MAGNESIA CRUCIBLES AND BRICKS. 131 
 
 native carbonate of magnesia, which formerly cost 10/. 
 the ton, may now be obtained at the price of 21. 15s. 
 delivered at Marseilles or Dunkirk. Calcination at the 
 place where the carbonate is obtained may still further 
 reduce its price. 1 The following is M. Caron's process 
 for its agglomeration, which may be employed by the 
 chemist for the ready preparation of refractory vessels 
 of all forms ; by the physicist to obtain pencils for oxy- 
 hydrogen lighting purposes ; and also by the manufac. 
 turer, to replace, in some cases, fire-bricks which have 
 become unsuitable for carrying out different processes 
 of heating. 
 
 The magnesia which he employs comes from the island 
 of Elba, where it is found in considerable quantities as a 
 native carbonate, white, very compact, and of great hard- 
 ness. This carbonate contains traces of lime, silica, and 
 iron ; it is, besides, interspersed sometimes with serpentine 
 and large plates of silica, which would diminish the in- 
 fusibility of the substance, and render it especially unfit 
 for oxyhydrogen illumination if their removal is neglected. 
 These plates are, however, easily recognised, and may 
 be readily separated, even in working on the large scale. 
 In the case of refractory bricks, the presence of a small 
 quantity of these foreign bodies would, at the most, give 
 rise, under the influence of the highest temperatures, to a 
 slight vitrification, offering no serious inconvenience. 
 
 Before powdering the carbonate, it is advisable to bake 
 it at the temperature necessary for the expulsion of the 
 carbonic acid ; the material then becomes very friable, 
 and may be pulverised more easily. It is then possible to 
 separate the serpentine and silica, which do not become 
 friable under the influence of heat. This preliminary 
 treatment does not permit of the agglomeration of the 
 magnesia, and, even were this difficulty to be overcome, a 
 temperature higher than that of the original calcination 
 would cause an enormous contraction, producing fissures 
 and alterations in shape which would interfere with the 
 
 1 This preliminary calcination requires less heat than burning lime, and 
 diminishes the weight of the carbonate one-half. 
 
 K 2 
 
132 MAGNESIA CRUCIBLES AND BRICKS. 
 
 use of this substance. It is therefore indispensable, before 
 moulding the magnesia, to submit it to a very intense heat, 
 at least equal to that which it is intended to support sub- 
 sequently. 
 
 Thus calcined it is not plastic, its appearance is sandy, 
 and compression does not cause it to acquire any cohesion ; 
 a mixture of magnesia, less calcined, imparts to it this 
 quality. The quantity of the latter to be added necessarily 
 varies with the degree of calcination of the two magnesias ; 
 it is scarcely one-sixth of the weight of that which has 
 been exposed to the temperature of melting steel. It only 
 now remains to moisten it with 10 or 15 per cent, of 
 its weight of water, and strongly compress it in iron 
 moulds, as adopted in making artificial fuel. The brick 
 produced in this operation hardens on drying in the air, 
 and becomes still more resisting when it is subsequently 
 calcined at a red heat. The same process would appear 
 practicable, varying the form of the moulds, for obtaining 
 crucibles of great capacity ; but compression is difficult in 
 large masses, as well as when the moulds have a large 
 surface, as the magnesia adheres strongly to the sides. 
 Although M. Caron has been able to obtain small crucibles 
 for the laboratory, he does not consider this process 
 adapted to make the large crucibles employed in steel 
 melting. In this and other cases it is preferable to ag- 
 glomerate the magnesia in the wet way. 
 
 To endow magnesia with a sort of plasticity, advantage 
 is taken of a property of this earth pointed out in ' Ber- 
 zelius's Chemistry.' When strongly calcined and then 
 moistened, it hardens in drying. This fact is doubtless 
 due to a hydration which takes place without sensible in- 
 crease of temperature. When solidified in this manner 
 the magnesia only loses the assimilated water at a high 
 temperature. Then calcination not only does not disin- 
 tegrate it, but, on the contrary, confers upon it a hardness 
 and resistance comparable to those of ordinary crucibles 
 after their baking. This being understood, the appli- 
 cation of this fact is obvious. Thus, the magnesia to be 
 employed in the manufacture of crucibles should only be 
 
PLATINUM CRUCIBLES. 133 
 
 moistened, moulded into shape, dried, and then ignited. 
 For the construction of steel-melting furnaces, a paste of 
 moistened magnesia should be plastered over the walls ; it 
 will become ignited in due course without any particular 
 precautions being taken. It sometimes happens, however, 
 either owing to the magnesia being too much or too little 
 hydrated, or owing to its containing siliceous matter, that 
 the vessels before or after firing do not possess quite the 
 desirable solidity ; they should then, to acquire this, be 
 simply moistened in a cold saturated solution of boracic 
 acid, dried, and then fired as before. This operation does 
 not render the magnesia more fusible ; it only causes the 
 grains of the substance to cohere more strongly together. 
 
 Very pure, strongly calcined, and finely pulverised 
 magnesia may be employed in the form of paste (barbotine), 
 and yields the most delicate and translucent crucibles, as 
 well as the sharpest and most complicated impressions. 
 
 MALLEABLE IRON CRUCIBLES are often very serviceable in 
 assays of fusibility, and of certain selenides and sulphides, 
 also in assays of galena or ordinary lead ore. They are 
 either made of hammered sheet iron, or by plugging up 
 small iron tubes, such as gun-barrels, &c. The latter are pre- 
 ferable, because thick solid crucibles can be used a number 
 of times, whilst the others are necessarily very thin and 
 can be used only once. Whenever iron crucibles are em- 
 ployed at a very high temperature, they must be placed 
 in those of earthenware, which protect them from the 
 oxidising action of the air ; but when they are not heated 
 above the temperature of a copper assay, they may be used 
 naked, if tolerably thick. 
 
 For assays at the above temperature, cast-iron crucibles 
 may be employed with advantage, instead of wrought-iron, 
 because they are very nearly as good, and much less 
 expensive. 
 
 PLATINUM CRUCIBLES. Platinum crucibles are invalu- 
 able in a laboratory. Few pieces of apparatus are used 
 so frequently by the chemist. Their chief use is in the 
 ignition of precipitates and the decomposition of siliceous 
 minerals by fusion with alkaline carbonates. They are 
 
134 PLATINUM CRUCIBLES. 
 
 preferable to porcelain, as not being fragile and being 
 more readily heated to redness over the gas or spirit flame. 
 Their most convenient size is 1^ inch high and 1J inch 
 wide at the top. 
 
 In employing a crucible for the incineration of filters 
 in quantitative assays by the wet way, it sometimes happens 
 (as, for instance, with chloride of silver or sulphate of lead) 
 that the employment of platinum is inadmissible. In these 
 cases thin porcelain crucibles must be used. The analyst 
 will, however, frequently experience difficulty, owing to 
 the extreme slowness w^ith which, in many cases, the last 
 portions of the carbon of a filter are consumed when 
 ignited in a porcelain crucible. It does not appear, how- 
 ever, that the following simple method of obviating the 
 difficulty, as practised in the laboratory of Professor 
 Sheerer, in Freiburg, has ever received the publicity which 
 it deserves. Whenever a filter upon which a substance 
 capable of injuring platinum has been collected, has to be 
 incinerated, the porcelain crucible or capsule in which the 
 process is to be conducted should be placed within a vessel 
 of platinum of similar form, and the whole ignited in the 
 usual way. Whether the greatly accelerated rapidity of 
 combustion of the carbon which ensues depends upon a 
 more equal distribution of heat brought about by the 
 greater conducting power of the metal an explanation 
 which is current for the somewhat analogous case of copper- 
 coated glass flasks or whether, as seems probable, the 
 power of the porcelain vessel to absorb heat be really 
 increased by the interposition of the platinum ; whether 
 both these causes be of influence, or the result depends 
 upon another less apparent reason ; or, finally, whether 
 vessels of some other metal would not be preferable to 
 those of platinum, are questions which are open to dis- 
 cussion. 
 
 Fresenius gives the following excellent directions as to 
 the preservation of platinum crucibles. The analyst should ; 
 acquire the habit of cleaning and polishing the platinum 
 crucible always after using it. This should be done, as 
 recommended by Berzelius, by friction with moist sea-sand, 
 
PLATINUM CRUCIBLES; 135 
 
 whose grains are all round, and do not scratch. The 
 writer has found this method to answer extremely well. 
 The sand is rubbed on with the finger, and the desired 
 effect is produced in a few minutes. The adoption of this 
 habit is attended with the pleasure of always working with 
 a bright crucible, and the profit of prolonging its existence. 
 This mode of cleaning is all the more necessary when one 
 ignites over gas-lamps, since at this high temperature 
 crucibles soon acquire a grey coating, which arises from 
 a superficial loosening of the platinum. A little scouring 
 with sea-sand readily removes the appearance in question, 
 without causing any notable diminution in the weight of 
 the crucible. 
 
 The ordinary Bunsen burner is known to act upon the 
 surface of platinum vessels brought into contact with the 
 inner line of the flame ; the metal loses its polish, becom- 
 ing superficially porous and spongy, and requires the use 
 of the burnisher to bring it back to its original state. This 
 alteration of the surface Mr. Dexter has found to be at- 
 tended with a change of weight, so that for some years he 
 has used a lamp of different construction for the heating 
 of platinum crucibles in analytical operations. Such a 
 lamp may be made by removing the air-tube of a common 
 Bunsen lamp, and putting in its place a somewhat longer 
 one of glass or iron of about 12 millimetres internal dia- 
 meter. The gas jet should have a single circular aperture, 
 and be in proper proportion to the diameter of the tube, 
 which may be held in any of the ordinary clamp supports. 
 The tube being raised sufficiently above the jet to allow 
 free entrance of air, and a full stream of gas let on, a ' roar- 
 ing ' flame is produced, of which the interior blue cone is 
 pointed, sharply defined, and extends only about half an 
 inch from the top of the tube. A polished platinum sur- 
 face is not acted upon by this flame, provided it be not 
 brought into contact with the interior cone. In the 
 Bunsen burner, as usually made, the supply of air depends 
 upon the diameter of the tube, the holes at its base being 
 more than sufficient to supply the draught. With the 
 wider tube it is necessary to limit the admission of air by 
 
130 -PLATINUM CRUCIBLES. 
 
 depressing the tube upon the lamp when the force of 
 the gas is diminished. Otherwise the proportion becomes 
 such that an explosive mixture is formed ; for this reason 
 it is more convenient to use an arrangement in which the 
 access of air can be regulated by an exterior tube sliding 
 obliquely downward over the air-apertures. The gas jet 
 should be on a level with the top of these apertures, which 
 must be much larger than those of the ordinary Bunsen's 
 burner. On account of the liability to explode and burn 
 at the jet inside, the lamp is not well adapted for ordinary 
 use ; but for ignition of crucibles, working of glass, &c., 
 it has proved efficient and practical. 
 
 In connection with some sensible remarks upon the 
 before-mentioned use of sand in cleaning platinum cru- 
 cibles, Erdmann explains in the following way the cause 
 of this grey coating which forms upon platinum crucibles 
 whenever they are ignited in the flame of Bunsen's gas- 
 burner. This coating has given rise to much annoyance 
 and solicitude among chemists. Indeed, it has often been 
 asserted that the use of Bunsen's burner is unadvisable in 
 quantitative analysis, since by means of it the weight of 
 platinum crucibles is altered and the crucibles themselves 
 injured. The coating is produced most rapidly when the 
 crucible is placed in the inner cone of the flame, and the 
 more readily in proportion as the pressure under which 
 the gas is burned is higher. Having found it advantageous 
 to maintain, by means of a special small gas-holder, a 
 pressure of four or five inches upon the gas used in his own 
 laboratory, Erdmann observed that the strong gas-flame 
 thus afforded immediately occasioned the formation of a 
 dull ring upon the polished metal placed in the inner 
 flame, this ring being especially conspicuous when the 
 crucible becomes red-hot ; it increased continually, so 
 that after long-continued ignition the whole of the bottom 
 of the crucible was found to be grey and with its lustre 
 dimmed. 
 
 The ring is caused neither by sulphur, as some have 
 believed, nor by a coating of inorganic matter, but is simply 
 a superficial loosening of the texture of the platinum, 
 
PRESERVATION OP PLATINUM CRUCIBLES. 137 
 
 in consequence of the strong heat, whence it first of all 
 appears in the hottest part of the flame. In consequence 
 of the serious damage which the gas furnace causes, many 
 chemists now discard gas and ignite platinum crucibles 
 over specially constructed spirit lamps. 
 
 In conjunction with Pettenkofer, Erdmann instituted 
 several experiments, which have left but little doubt that 
 the phenomenon depends upon a molecular alteration of 
 the surface of the metal. If a weighed polished crucible 
 be ignited for a long time over Bunsen's lamp, the position 
 of the crucible being changed from time to time, in order- 
 that the greatest possible portion of its surface shall be 
 covered with the grey coating, and its weight be then de- 
 termined anew, it will be found that this has not increased. 
 The coating cannot be removed either by melting with 
 bisulphate of potash or with carbonate of soda. It dis- 
 appears, however, when the metal is polished with sand ; 
 the loss of weight which the crucible undergoes being 
 exceedingly insignificant, a crucible weighing 25 grammes 
 having lost hardly half a milligramme. When the grey 
 coating of the crucible is examined under the microscope, 
 it may be clearly seen that the metal has acquired a rough, 
 almost warty, surface, which disappears when it is polished 
 with sand. Platinum wires, which are frequently ignited 
 in the gas-flame for example, the triangles which are 
 used to support crucibles become, as it is known, grey 
 and brittle. Under the microscope they exhibit a mul- 
 titude of fine longitudinal cracks, which, as the original 
 superficial alteration penetrates deeper, become more open, 
 or, as it were, spongy, until finally the wire breaks. 
 
 If such wire is strongly and perseveringly rubbed with 
 sand, the cracks disappear, ana the wire becomes smooth 
 and polished ; for the grains of sand, acting like burnishers, 
 restore the original tenacity of the metal, very little of its 
 substance being rubbed off meanwhile. The loosening 
 effect of a strong heat upon metals is beautifully exhibited 
 when silver is ignited in the gas-flame, a thick polished 
 sheet of silver immediately becoming dull white when thus 
 heated. Under the microscope the metal appears swollen 
 
138 CLEANING PLATINUM CRUCIBLES. 
 
 and warty. Where it has been exposed to the action of 
 the inner flame along its circumference, this warty con- 
 dition is visible to the naked eye. A stroke with the 
 burnishing stone, however, presses down the loosened par- 
 ticles, and reproduces the original polish. This peculiar 
 condition which the surface of silver assumes when it is 
 ignited, is well known to silversmiths ; it cannot be re- 
 placed by any etching with acids, and it must be remem- 
 bered that what is dull white in silver appears grey in 
 platinum. 
 
 If each commencement of this loosening is again de- 
 stroyed, the crucibles will be preserved unaltered, other- 
 wise they must gradually become brittle. Crucibles of 
 the alloy of platinum and iridium are altered, like those of 
 platinum, when they are ignited. It is, however, some- 
 what more difficult to reproduce the original polish of the 
 metal by means of sand, as might be expected, from the 
 greater hardness of the alloy. 
 
 The sand used should be well worn. When examined 
 under the microscope no grain of it should exhibit sharp 
 edges or corners ; all the angles should be obtuse. 
 
 If there are spots on the platinum crucibles which 
 cannot be removed by the sand without wearing away too 
 much of the metal, a little potassium bisulphate is fused 
 in the crucible, the fluid mass shaken about inside, allowed 
 to cool, and the crucible finally boiled with water. There 
 are two ways of cleaning crucibles soiled outside ; either 
 the crucible is placed in a larger one, and the interspace 
 filled with potassium bisulphate, which is then heated to 
 fusion, or the crucible is placed on a platinum-wire triangle 
 heated to redness, and then sprinkled over with powdered 
 potassium bisulphate. Instead of the bisulphate, borax 
 may be used. Never forget at last to polish the crucible 
 with sea-sand again. 
 
 A remarkably rapid and perfect method of cleaning 
 platinum apparatus consists in gently rubbing upon the 
 dirty metal a small lump of sodium-amalgam. Sodium has 
 the curious property of lending to mercury the power of 
 ' wetting ' platinum in so complete a manner that the 
 
CLEANING PLATINUM CRUCIBLES. 139 
 
 positive capillarity between platinum and an amalgam con- 
 taining even only one per cent, of sodium appears to be 
 as great as that between mercury and zinc, with this 
 important difference, however in the former case, the 
 4 wetted ' metal does not suffer the least trace of amalga- 
 mation. Even when foreign metals, such as lead, tin, zinc r 
 silver, are purposely added to the sodium -amalgam, the 
 platinum surface suffers no disintegration. 
 
 When the amalgam has been rubbed on with a cloth 
 until the whole surface is brilliantly metallic, water is ap- 
 plied which oxidises the sodium and allows the cohesion 
 of the mercury to assert itself. On wiping the mercury 
 off, the platinum surface is left in admirable condition for 
 the burnisher. 
 
 When the crucible is clean it is placed upon a clear 
 platinum-wire triangle, ignited, allowed to cool in the 
 desiccator, and weighed. This operation, though not in--, 
 dispensable, is still always advisable, that the weighing of; 
 the empty and the tilled crucible may be performed under 
 as nearly as possible the same circumstances. 
 
 In using platinum crucibles, it must be remembered 
 that certain substances must not be ignited in them. 
 Berzelius says that ' it is improper to ignite in platinum 
 vessels the caustic alkalies or the nitrates of any alkaline 
 base, such as lime, baryta, or strontia, because the affinity 
 of the alkali for platinum oxide causes a very considerable 
 oxidation of the metal ; and after the saline matter is re- 
 moved, the surface of the metal is found to be honey- 
 combed.' 
 
 The alkaline sulphides or the alkaline sulphates mixed 
 with charcoal are inadmissible, because the sulphides so- 
 formed attack platinum even more energetically than the 
 caustic alkalies ; so are metals whose fusing-point is lower 
 than that of platinum, because an alloy would be formed. 
 Gold, silver, and copper may be heated to dull redness in 
 platinurn vessels without danger ; but fused lead cannot 
 come in contact with platinum without destroying it. A 
 drop of fused lead, tin, zinc, or bismuth, placed on red-hot 
 platinum, always produces a hole. Neither can a phosphide 
 
140 SILVER CRUCIBLES NICKEL CRUCIBLES. 
 
 or phosphoric acid mixed with charcoal be ignited in 
 vessels of platinum, because a platinum phosphide is pro- 
 duced, which is an exceedingly brittle compound. 
 
 In analyses by the wet method, nitro-hydrochloric 
 acid (aqua regia), even when very dilute, must not be 
 allowed to come in contact with platinum, and, as a 
 general rule, liquids containing either free chlorine, bro- 
 mine, or iodine must not be boiled in platinum capsules. 
 
 SILVER CRUCIBLES can only be used at temperatures 
 below full redness. They are not affected by caustic 
 alkalies, but must not come in contact with sulphur or 
 be heated over coke, coal, gas, or other fuel containing 
 sulphur. 
 
 NICKEL CRUCIBLES. Recently nickel has been intro- 
 duced as a material for crucibles. It is well known that 
 pure nickel is one of the toughest of all the metals, 
 .and that it fuses only at very high temperatures. It lias 
 a fine grain, takes a high polish, and is very compact 
 and unalterable. These qualities have led to its being 
 employed for crucibles and evaporating dishes. Mr. 
 Wanklyn has published some notes on the behaviour of 
 these vessels in his laboratory. He finds that for many 
 purposes crucibles of pure nickel are quite as serviceable 
 as platinum crucibles, and they are much cheaper, costing 
 only about one-tenth as much as platinum. They stand 
 the action of alkalies remarkably well there was no 
 alteration in the weight of the crucible after caustic pot- 
 ash had been fused in it. Hydrochloric acid in the cold, 
 whether dilute or concentrated, may be used to clean out 
 these crucibles and no alteration in weight is the result. 
 Cold oil of vitriol is likewise without action ; but con- 
 centrated nitric acid attacks them, causing rapid loss of 
 weight. Nickel dishes are especially useful for taking 
 water-residues and milk- solids, and indeed for these pur- 
 poses nickel is not inferior to platinum. 
 
 Mr. Bertram Blount considers that the uses of nickel 
 for chemical purposes are confined to 
 
 (i.) Dry ignitions (i.e. not fusions) in an oxidising 
 flame, provided the heat be not too intense. 
 
CUPELS. 141 
 
 (ii.) Fusions with caustic alkalies. 
 From which may probably be deduced 
 
 (iii.) Fusions with barium hydrate. 
 
 (iv.) Fusions with alkaline carbonates. 
 
 The inability of nickel to stand high temperatures, in 
 contact with a reducing flame, limits its employment con- 
 siderably, and demands caution in dealing with it. 
 
 CUPELS. These are vessels in which the operation 
 termed cupellation is carried on. They must be made of 
 such substances as are not acted upon by certain fused 
 oxides, as those of lead or bismuth, and their texture has 
 to be sufficiently loose to allow of the oxides penetrating 
 their substance readily, and yet be sufficiently strong to 
 bear handling without breaking. 
 
 There are several substances of which cupels might be 
 made, which will fulfil all these conditions, but only one 
 is in general use, viz. the ash of burnt bones. This consists 
 principally of calcium phosphate, with a little calcium car- 
 bonate and fluoride. Berzelius found that bones of oxen 
 contained 57 parts of calcium phosphate for every 3'8 
 parts of calcium carbonate, whilst, according to Barros, 
 sheep bones contain 80 parts of calcium phosphate to 19 
 parts of calcium carbonate. When bones are burnt whole 
 they likewise contain mineral matter derived from the 
 cartilage, such as alkaline sulphates and carbonates. The 
 greater part of the calcium carbonate is likewise converted 
 into caustic lime. 
 
 The bones of sheep and horses are best for cupels. In 
 getting rid of the organic matter, it is advisable to boil 
 them repeatedly in water before burning them. This dis- 
 solves a great part of the organic matter. If the bones 
 are not rendered quite white by the first ignition, but con- 
 tain a little carbon, they should be ground up, moulded 
 into shape, and burned again. 
 
 Care should be taken not to heat the bone earth too 
 strongly. In this case the bones will have a smooth, glassy 
 fracture, and will not be sufficiently spongy or absorbent 
 to make good cupels. 
 
 When the bones are burnt white throughout, they must 
 
142 MANUFACTURE OP CUPELS. 
 
 be finely ground, sifted, and washed several times with 
 boiling distilled water till all soluble salts are removed. 
 The finest particles of the powdered bone earth will remain 
 longest suspended in the washing waters. This must be 
 allowed to settle separately, and should be reserved for 
 giving a final coating to the surface of the cupels ; this 
 coating acts, to a certain extent, like a fine filter, and may 
 be applied to all cupels, although the body of the cupel is 
 made of different materials. 
 
 For the body of the cupels, the bone-ash should be 
 -about as fine as wheat flour. If too coarse, litharge con- 
 taining silver will be absorbed into its pores, and will occa- 
 sion a loss of silver. 
 
 Cupels must neither crack nor alter in texture at a 
 white heat. It is very important that they should not 
 contain carbon, and therefore, in making them, the bone 
 earth must not, as sometimes recommended, be mixed with 
 beer, or water containing adhesive substances. Nothing 
 but pure water should be used, and the mixture should be 
 just sufficiently moist to adhere strongly when well pressed, 
 but not so moist as to adhere to the finger or the mould 
 employed to fashion the cupels. The mould (fig. 62) con- 
 FIG. 62. sists of three pieces : one a ring, b, having a 
 conical opening ; another a pestle, a, having 
 a hemispherical end fitting the larger opening 
 of the ring ; and the third, c, a piece of turned 
 metal, into which b fits ; c serves to form an 
 even bottom to the cupel. In order to mould 
 the cupels, proceed as follows : Place the ring 
 on the lower piece e, and fill it with the com- 
 position : then place the pestle upon it, and 
 force it down as much as possible : by this means the 
 moistened bone ash will become hardened, and take the 
 form of the pestle ; the latter must then be driven as 
 much as possible by repeated blows from a hammer, 
 until quite home. The surface of the cupel may then 
 have sifted over it a little of the very fine levigated bone- 
 ash, and the pestle hammered again on it. It is then to 
 be turned lightly round, so as to smooth the inner sur- 
 
MANUFACTURE OF CUPELS. 
 
 143 
 
 face of the cupel, and withdrawn : the cupel is removed 
 from the mould by a gentle pressure on the narrowest end. 
 When in this state, the cupel must be dried gently by a 
 stove ; and lastly, ignited in a muffle, to expel all moisture. 
 It is then ready for use. 
 
 There are two or three points to attend to in manufac- 
 turing the best cupels. First, the powdered bone-ash must 
 be of a certain degree of fineness ; secondly, the paste must 
 be neither too soft nor too dry ; and thirdly, the pressure 
 must be made with a certain degree of force. A coarse 
 powder, only slightly moistened and compressed, furnishes 
 cupels which are very porous, break on the least pressure, 
 and, as before mentioned, allow small globules of metal to 
 enter into their pores. 
 
 When, on the contrary, the powder is very fine, the 
 paste moist, and compressed strongly, the cupels have 
 much solidity, and are less porous ; the fine metal cannot 
 penetrate them, but the operation proceeds very slowly : 
 besides, the assay is likely to become dulled, and incapable 
 of proceeding without a much higher degree of tempera- 
 ture being employed. 
 
 Cupels for assaying silver bullion are sometimes made 
 of equal parts of bone-ash and beechwood-ash ; and for 
 assaying gold, 2 parts of beechwood-ash and 1 part of 
 bone-ash are used. The hemispherical cavity of both 
 these kinds are coated with the fine levigated powder of 
 bone-ash . 
 
 Beechwood-ash is preferred for the manufacture of 
 cupels on account of the larger proportion of phosphoric 
 acid it contains. 
 
 According to Hertwig, beechwood-ash contains in 100 
 parts : 
 
 Potassium carbonate 
 
 
 
 
 11-72 
 
 Sodium carbonate 
 
 
 
 
 12-37 
 
 Potassium sulphate 
 
 . 
 
 
 
 3-49 
 
 Calcium carbonate 
 
 
 
 
 49-54 
 
 Magnesium carbonat 
 
 :* 
 
 
 
 7-79 
 
 Calcium phosphate 
 
 
 
 
 3-32 
 
 Magnesium 
 
 
 
 
 2-92 
 
 Iron 
 
 
 
 
 0-76 
 
 Aluminium ,, 
 
 
 
 
 1-51 
 
 Manganese 
 Silica 
 
 
 
 
 1-59 
 2-46 
 
144 PYROMETRY. 
 
 SCORIFIERS. A scorifier (fig. 63) is a vessel made much 
 in the shape of a cupel, but of crucible earth. The proper 
 use both of cupels and scorifiers will be explained under 
 the head of silver- assaying. 
 
 METHODS OF MEASURING THE HEAT OF FURNACES. As 
 much of the accuracy of an assay depends on the tempera- 
 ture at which it is made, and as the temperature required 
 
 FIG. 63. 
 
 varies with each metal, it is very desirable to possess some 
 means of ascertaining the heat of the furnace more accu- 
 rately than by the eye. Many persons have devised 
 instruments, called pyrometers, for this purpose ; the 
 earliest being those of Mr. Wedgwood and the late Pro- 
 fessor Daniel, of King's College. 
 
 We shall not give a description of Wedgwood's pyro- 
 meter, as its indications are inaccurate, from the fact that 
 the clay cylinders, whose contraction serves to measure 
 the temperature, will contract as much as by the long con- 
 tinuance of a low heat as by the short continuance of a 
 high one. Hence the degrees of heat measured by Wedg- 
 wood's pyrometer have been enormously exaggerated. It 
 was long since noticed that it did not produce comparable 
 effects; and this was supposed to proceed wholly from 
 the impossibility of obtaining clay perfectly alike for each 
 experiment. 
 
 Daniel's pyrometer is composed of a rod of platinum 
 simply laid in a groove made of refractory clay, and baked 
 in the highest degree of heat. This rod rests at one end 
 
DANIELS PYROMETER. 
 
 145 
 
 on the edge which terminates the groove, and at the other 
 on a lever with two arms, the larger of which forms a 
 needle on a graduated arc of a circle ; so that the removal 
 of this needle from its position marks the additional length 
 which the metal acquires by the heat. The following is 
 Daniel's description of his pyrometer : ' It consists of two 
 parts (see fig. 64), which may be distinguished as the 
 
 The register is a solid bar of black- 
 
 register and the scale. 
 
 FIG. 64. 
 
 J 
 
 lead or earthenware highly baked. In this a hole is drilled, 
 into which a bar of any metal, a, six inches long, may be 
 dropped, and which will then rest upon its solid end. A 
 cylindrical piece of porcelain, , called the index, is then 
 placed upon the top of the bar, and confined in its place 
 by a ring or strap of platinum passing round the top of 
 the register, which is partly cut away at the top, and 
 tightened by a wedge of porcelain. When such an arrange- 
 ment is exposed to a high temperature, it is obvious that 
 the expansion of the metallic bar will force the index for- 
 ward to the amount of the excess of its expansion over that 
 of the black-lead, and that when again cool it will be left 
 at the point of greatest elongation. What is now required 
 is the measurement of the distance which the index has 
 been thrust forward from its first position, and this, 
 
 L 
 
146 DANIEL S PYROMETER. 
 
 though in any case but small, may be effected with great 
 precision by means of the scale c.' * 
 
 This is independent of the register, and consists of two 
 rules of brass accurately joined together at a right angle 
 by their edges, and fitting square upon the two sides of 
 the black-lead bar. At one end of this double rule a small 
 plate of brass projects at a right angle, which may be 
 brought down upon the shoulder of the register formed by 
 the notch cut away for the reception of the index. A 
 movable arm is attached to this frame, turning at its fixed 
 extremity on a centre, and at its other carrying the arc of 
 a circle, whose radius is exactly five inches, accurately 
 divided into degrees, and thirds of a degree. Upon this 
 arm, at the centre of the circle, another lighter arm is made 
 to turn, one end of which carries a vernier with it, which 
 moves upon the face of the arc, and subdivides the former 
 graduation into minutes of a degree ; the other end crosses 
 the centre, and terminates in an obtuse steel point, turned 
 inwards at a right angle. When an observation is to be 
 made, a bar of platinum or malleable iron is placed in the 
 cavity of the register ; the index is to be pressed down upon 
 it, and firmly fixed in its place by the platinum strap and 
 porcelain wedge. The scale is then to be applied by care- 
 fully adjusting the brass rule to the sides of the register, 
 and fixing it by pressing the cross piece upon the shoulder, 
 and placing the movable arm so that the steel part of the 
 radius may drop into a small cavity made for its reception, 
 and coinciding with the axis of the metallic bar. The minute 
 of the degree must then be noted which the vernier indi- 
 cates upon the arc. A similar observation must be made 
 after the register has been exposed to the increased tem- 
 perature which it is designed to measure, and again cooled, 
 and it will be found that the vernier has been moved 
 forward a certain number of degrees or minutes. The 
 scale of this pyrometer is readily connected with that of 
 the thermometer by immersing the register in boiling 
 mercury, whose temperature is as constant as that of boil- 
 ing water, and has been accurately determined by the 
 
 * Daniel's ' Chemical Philosophy,' p. 111. 
 
DANIEL S PYROMETER, 
 
 147 
 
 thermometer. The amount of expansion for a known 
 number of degrees is thus determined, and the value of 
 all other expansions may be considered as proportionate. 
 
 By Daniel's pyrometer the melting-point of cast iron 
 has been ascertained to be 2,786, and the highest tem- 
 perature of a good wind furnace 3,300 Fahrenheit 
 points which were estimated by Mr. Wedgwood at 17,977 
 and 21,877 respectively. 
 
 The following is a list of the melting-points of some of 
 the metals as ascertained by Professor Daniel ; and it is 
 obvious that in an assay of each particular metal the tem- 
 perature employed must exceed by a considerable number 
 of degrees its melting-point. The table is, therefore, very 
 useful. 
 
 Fahr. 
 
 Tin melts at - - - - - - - 4-22 
 
 Cadmium 
 
 
 
 
 
 
 
 
 442 
 
 Bismuth 
 
 
 
 
 
 
 
 
 497 
 
 Lead . 
 
 
 
 
 
 
 
 
 612 
 
 Zinc . 
 
 
 
 
 
 
 
 
 773 
 
 Silver 
 
 
 
 
 
 
 
 
 1860 
 
 Copper 
 
 
 
 
 
 
 
 
 1996 
 
 Gold . 
 
 
 
 
 
 
 
 
 2016 
 
 Cast iron 
 
 
 
 
 
 
 
 
 2786 
 
 Cobalt and nickel are rather less fusible than iron. 
 
 Mr. S. Wilson* has described an ingenious process of 
 measuring high temperatures. He exposes a given weight 
 of platinum or Stourbridge clay to the action of the heat 
 which is to be measured, and then quenches it in a definite 
 weight of water at a certain temperature. Thus, if the 
 piece of platinum weigh 1,000 grains and the water 2,000 
 grains at 60 F., and should the heated platinum when 
 dropped into the water raise its temperature to 90, then 
 90- 60 =30 ; which, multiplied by 2 (because the weight 
 of the water is twice that of the platinum), gives 60 
 the temperature to which a weight of water equal to the 
 platinum would have been raised. To convert this into 
 Fahrenheit degrees we must multiply by 31J, which is the 
 specific heat of water as compared with platinum, that of 
 the latter being 1. Therefore 60 x 31 J 1875, which 
 will be the temperature of the furnace. 
 
 Philosophical Magazine,' ser. iv. vol. iv. p. 157. 
 
 L2 
 
148 
 
 COMPARISON OF HIGH TEMPERATURES. 
 
 FIG. 65. 
 
 One or two other methods of measuring high tempera- 
 tures applicable to special cases may here be mentioned. 
 
 Mr. C. W. Siemens, C.E., F.E.S., has invented an inge- 
 nious pyrometer, the principle of which is, that as the 
 electrical conductivity of platinum, iron, and 
 other metals decreases as they rise in tem- 
 perature, their increase of resistance to the 
 passage of the current is a measure of the heat 
 to which the metals are subjected. 
 
 The principle of construction may be ex- 
 plained by the aid of fig. 65, in which F A B 
 is a tube of pipe-clay, and the length between 
 the projections A and B has a screw-shaped 
 spiral groove cut on its outer surface ; the 
 length of this part of the tube is about 3 inches. 
 A spiral of fine platinum wire lies in the groove, 
 each turn of the platinum spiral being thus pro- 
 tected from lying in contact with its neighbour 
 by the projecting edges of the groove, by which 
 plan of insulation the current is forced to pass 
 through the whole length of the fine wire. D 
 is a little platinum clam, connected with one 
 pole of the battery, and the position of this 
 clam on the spiral regulates the length of pla- 
 tinum wire through which the current shall 
 pass. By this plan of adjustment all the pyro- 
 meters constructed by Mr. Siemens are made 
 to agree with each other. 
 
 At F the ends of the thin platinum wire are connected 
 with very thick platinum wire, and higher up, near E r 
 where the heat of the furnace is less felt, the thick 
 platinum wires are connected with thick copper wires, 
 shown at P ; from E to F these connecting wires are pro- 
 tected by clay pipes, as shown in the cut. 
 
 When this arrangement has to be used, the whole of it 
 is dropped into a thick metal pipe made of iron, copper, 
 or platinum, according to the heat of the furnace to be 
 tested. The lower end of this outer pipe is shown at KM, 
 and when it is used the spiral A B lies inside it at N M. At 
 
COMPARISON OF HIGH TEMPERATURES. 149 
 
 R there is a very thick collar of metal in which the heat 
 accumulates, and this prevents the cooling action of the 
 length K R (most of which does not enter the furnace) from 
 interfering with the accuracy of the indications. The 
 ends of the wires p are connected with a suitable and very 
 delicate electrical apparatus, by which the increasing elec- 
 trical resistance of the hot spiral is measured. 
 
 A good plan for comparing the temperatures of two 
 furnaces is to prepare alloys of platinum and gold, con- 
 taining definite quantities, say 5, 10, 15, 20 per cent. &c., 
 of gold. These fuse at intermediate temperatures between 
 gold and platinum. By placing small angular chips of 
 these alloys separately in muffles, and noticing which are 
 melted, which softened only, and which resist the action 
 of the heat, an idea of the power of the furnace is obtained. 
 In this way the amount of heat required to perform any 
 operation may be registered for future reference, by sim- 
 ply recording that it was sufficient just to melt, say, a 20 
 gold 80 platinum alloy. 
 
150 
 
 CHAPTEE V. 
 
 FUEL : ITS ASSAY AND ANALYSIS. 
 
 BEFORE treating of the assay of metals and metalliferous- 
 ores, it is advisable to devote some space to the important 
 subject of fuel. The substances employed as fuel, although 
 all of vegetable origin, are derived either from the veget- 
 able kingdom (wood), or from the mineral kingdom (peat, 
 brown coal, coal, anthracite). These natural fuels can 
 be converted into artificial fuels by heating them more or 
 less out of contact with the air (charcoal, turf-charcoal, 
 coke). 
 
 The essential elements of combustible matters are car- 
 bon, oxygen, and hydrogen ; nitrogen being present some- 
 times, but only in small proportions. These constitute 
 the organic part ; various salts and silica constitute the 
 inorganic part, or ash. The valuable constituents of fuel, 
 on which its calorific and reducing powers depend, are 
 the carbon and hydrogen, and it is upon the combustion 
 or union of these elements with oxygen to form carbonic 
 acid and water that the heating effect of the fuel depends. 
 
 The more oxygen a fuel contains, the less carbon and 
 combustible gases it will yield, and the more hydrogen, 
 the more combustible gases. 
 
 The proportion of hydrogen to oxygen in wood . is 1 
 
 turf . 1 
 
 fossil wood 1 
 
 coal . 1 
 
 anthracite 1 
 
 6 
 4 
 
 2-3 
 1 
 
 The more oxygen, the less carbon the fuel contains, 
 thus : 
 
 Anthracite contains about 90 per cent, carbon 
 Coal 80 
 
 Brown coal 70 
 
 Fossil wood and turf 60 
 
 Wood 60 
 
ASSAY OF FUEL. 151 
 
 The more carbon a fuel contains, the greater heat it pro- 
 duces, and the more difficult it is to ignite. 
 
 The greater the amount of hydrogen in a fuel, the 
 more inflammable it will be, and the larger flame it gives, 
 the hydrogen being evolved below a red heat. But the more 
 carbon present the less flame. These differences are shown 
 in a blazing fire and a glowing fire. In a flame the hottest 
 part is at the periphery, whilst in a glowing fire the greatest 
 heat is in the immediate contact of the burning surface. 
 
 An elementary analysis of coal teaches little with 
 regard to the nature or practical value of the combustible. 
 A proximate analysis, on the contrary, enables us to learn 
 something in regard to the real nature of the coal. The 
 moisture and ash are not only diluents of the fuel, but 
 are in themselves obstacles to its effectiveness ; the vapo- 
 risation of the moisture causes a serious loss of heat, whilst 
 the ashes, by hindering complete combustion and by the 
 heat they contain when dropped through the grate, con- 
 stitute another loss. By furthermore determining the 
 total amount of volatile matter we learn both the per- 
 centage of coke in the fuel and the amount of carbon 
 (fixed combustible) and bitumen (volatile combustible 
 matter). Although neither of these two products can 
 be considered as simple chemical . compounds, it is never- 
 theless of the utmost practical importance to know these 
 two quantities, because of the great value of coke and gas 
 in manufactures. 
 
 The assay of fuel comprises the following examina- 
 tions : 
 
 1. The examination of the external appearance of the 
 fuel, its porosity or compactness, its fracture, the size and 
 shape of the pieces composing it. 
 
 2. Estimation of the adhering water. 
 
 3. Estimation of the specific gravity. 
 
 4. Estimation of the absolute heating power. 
 
 5. Estimation of the specific heating power. 
 
 6. Estimation of the pyrometric heating power. 
 
 7. Estimation of the volatile products of carbonisa- 
 tion. 
 
:152 ASSAY OF FUEL. 
 
 8. Examination of the coke or charcoal left behind on 
 carbonisation, both with regard to quality and quantity. 
 
 9. Estimation of the amount of ash, and its composi- 
 tion. 
 
 10. Estimation of the amount of sulphur. 
 
 11. Examination of any other peculiarity which may 
 be noticed during the burning or carbonisation of the 
 fuel. 
 
 1. EXTERNAL APPEARANCE OF THE FUEL, ITS POROSITY, 
 COMPACTNESS, FRACTURE, SIZE, AND SHAPE OF PIECES. From 
 the outward appearance of a fuel, its cleavage, and an 
 examination of the embedded earthy matter, iron pyrites, 
 gypsum, &c., its applicability to any special purpose may 
 be judged. Its degree of inflammability, together with 
 the pressure of blast which it will bear in the furnace, 
 partly depend on the more or less compactness of the fuel. 
 iThe amount of loss which it will suffer in transport de- 
 pends upon its friability. Playfair and De la Beche * 
 estimated the amount of this loss in coal by rotating 
 in a barrel different qualities of coal for the same time. 
 The powder produced was separated and weighed, and in 
 this way the friability or cohesion of a fuel could be ex- 
 pressed in percentages. Schrotter f made the same experi- 
 ments with brown coal. 
 
 The size and form of the pieces composing the fuel is 
 important, as on this depends the space occupied in its 
 stowage an important point for steam-vessels. This 
 space cannot be calculated from its specific gravity, but 
 must be ascertained by direct measurement. The space 
 occupied will be smallest when the form of the lumps is 
 cubical. 
 
 2. ESTIMATION OF THE ADHERING WATER. The water 
 contained in a fuel exerts great influence on its heating 
 power. It not only increases its bulk, but it acts in- 
 juriously by abstracting a certain quantity of heat required 
 for its evaporation, and it also causes imperfect combus- 
 
 * Dingl. ex. 212, 262; cxiv. 346. Liebig's * Jahresber.,' 1847-1848, 
 p. 1117; 1849, p. 708. 
 
 t Wien. Akad. Ber. 1849, Nov. and Dec. p. 240. Liebig's ' Jahresber., 
 1849, p. 709. 
 
ASSAY OF FUEL. 15U 
 
 tion. For this reason, wood, turf, and brown coal never 
 give so high a temperature as coal, anthracite, and coke. 
 
 The estimation of the adhering water is effected by 
 drying a certain weight of the pounded fuel in a water- 
 bath at 212 F., or in an air-bath at 220. It may also be 
 ascertained by placing a certain weight of the powdered 
 fuel in a glass tube, heating to 212, and passing over it 
 air dried by means of chloride of calcium, till the fuel no 
 longer loses weight. The amount of water which the 
 dried fuel will absorb from the atmosphere in twenty-four 
 hours should also be estimated, in order to ascertain its 
 hygroscopic qualities. 
 
 3. ESTIMATION OF THE SPECIFIC GRAVITY. The specific 
 gravity of a fuel depends on its density and the amount 
 of ash, and it appears also to be in proportion to its greater 
 or less inflammability. Of two equal volumes of carbonised 
 fuel, the one will produce the greatest heating effect which 
 has the greatest specific gravity, provided the density is 
 not produced by mineral constituents. 
 
 The estimation of the specific gravity is difficult, and 
 sometimes uncertain, owing to the cleavage of the fuel, 
 and the entanglement of air in its pores. The best way 
 of obviating this difficulty is as follows : 
 
 Coarse fragments, freed by means of a sieve from all 
 small particles, and averaging l-10th c.c. in volume, are 
 introduced into a fifty-gramme flask provided with a 
 thermometer stopper. The constants for this flask for 
 temperatures varying from 50 to 80 F. are previously 
 carefully estimated. 
 
 The true sp. gr. corresponds to the coal perfectly 
 soaked, so that all its pores are filled with water. That 
 requires, on the average, 12 hours, permitting two es- 
 timations per day, one in the morning, another in the 
 evening. 
 
 That this precaution is important may be seen from 
 the following example : a sample of coal gave the sp. gr. 
 1-309 at 64 F., immediately after filling the flask with 
 water : after about 12 hours' soaking, the sp. gr. had 
 increased to 1-328, for the same temperature. According 
 
154 ASSAY OF FUEL. 
 
 to this latter estimation, a cubic foot of this coal would 
 weigh 82-76 Ibs. ; according to the former, only 81*58, or 
 1-18 Ibs. less. This shows a considerable degree of po- 
 rosity of the coal, and indicates the absurdity of giving 
 the weight in pounds of a cubic foot of coal with four 
 decimals, although no statement in regard to temperature 
 or time of weighing is made. 
 
 4. ESTIMATION OF THE ABSOLUTE HEATING POWER. The 
 value of a fuel for any purpose depends chiefly on its 
 price and the quantity required for that purpose. The 
 quantity required depends on the heating power possessed 
 by a certain weight of fuel (its absolute heating power) or 
 that possessed by a certain volume (its specific heating 
 power). 
 
 The less oxygen, ash, and water the fuel contains, the 
 greater its heating power will be, and this will also increase 
 in proportion to the carbon and hydrogen present. 
 
 Whether the combustion is effected quickly or slowly, 
 the amount of heat produced will be the same, but the 
 degree of temperature attained will be very difficult. This 
 latter constitutes the pyrometric heating power. 
 
 The estimation of the absolute heating power of a fuel 
 may be effected 
 
 a. By heating a definite quantity of water from 32 F. 
 to 212 ; 
 
 . b. By ascertaining how much fuel is required to melt 
 a known weight of ice ; 
 
 c. By ascertaining how much water may be evaporated 
 by 1 Ib. of different kinds of fuel ; 
 
 d. By ascertaining how much the temperature of a 
 room increases by burning a certain weight of a fuel in a 
 stove. 
 
 e. By ascertaining the elementary composition of the 
 fuel, and calculating how much oxygen will be required 
 to convert the carbon and hydrogen into carbonic acid 
 and. water ; the quantity of heat produced will be in pro- 
 portion to the amount of oxygen consumed. 
 
 /. By Berthier's method. 
 g. By lire's method. 
 
BERTHIER'S METHOD FOE ASSAY OF FUEL. 155 
 
 According to Berthier, the most convenient method 
 for ascertaining the comparative calorific power of any 
 combustible matter is by means of litharge. He says : It 
 has been proved by the experiments of many philosophers 
 that the quantities of heat emitted by combustible sub- 
 stances are exactly proportioned to the amounts of oxygen 
 required for their complete combustion. Whence, after 
 the elementary constitution of any combustible is known, 
 its calorific power is easily estimated by calculation. 
 For instance, it is only necessary to ascertain the quantity 
 of oxygen absorbed in the conversion of all its carbon 
 into carbonic acid, and all its hydrogen into water, and 
 compare that quantity with that which is consumed in 
 burning a fuel whose calorific power is well ascertained. 
 Such a fuel is pure charcoal. 
 
 By adopting the principle just pointed out, it may be 
 conceived that, without knowing the composition of a 
 fuel, its heating power may be ascertained by estimating 
 the amount of oxygen it absorbs in burning. This can be 
 done in a very simple and expeditious manner, if not 
 exactly, at least with sufficient exactitude to afford very 
 useful results in practice. It is as follows : many metallic 
 oxides are reduced with such facility that when heated 
 with a combustible body, the latter burns completely, 
 without any of its elements escaping the action of the 
 oxygen of the oxide, if the operation be suitably per- 
 formed. The composition of the oxide being well known> 
 if the weight of the part reduced to the metallic state be 
 taken, the quantity of oxygen employed in the combustion 
 can be ascertained. In order to collect the metal and 
 separate it from the non-reduced mass, it, as well as its 
 oxide, must be fusible. Litharge fulfils these conditions, and 
 experiment has proved that it completely burns the greater 
 part of all ordinary fuels ; the only exceptions are some 
 very bituminous matters containing a large proportion of 
 volatile elements, a portion of which escapes before the 
 temperature is sufficiently high to allow the reduction to 
 take place. The experiment is made as follows : 10 grains 
 of the finely powdered or otherwise divided fuel is mixed 
 
156 BERTHIER'S METHOD FOR ASSAY OF FUEL. 
 
 with about 400 grains of litharge. The mixture is care- 
 fully placed in an earthen crucible, and covered with 200 
 grains more litharge. The crucible is then placed in the 
 fire and gradually heated. When the fusion is perfect, the 
 heat is urged for about ten minutes, in order that all the 
 lead may collect into a single button. The crucible is 
 then taken from the fire, cooled, broken, and the button 
 of lead weighed. Sometimes the button is livid, leafy, 
 and only slightly ductile ; in/ which case it has absorbed 
 a little litharge. This can be partially prevented by fusing 
 slowly, and adding a little borax. 
 
 Two assays, at least, ought to be made, and those re- 
 sults which differ more than a grain or two ought not to 
 be relied on. The purer the litharge, the better the re- 
 sult ; it ought to contain as little minium as possible. It 
 is an excellent plan to mix up the litharge of commerce 
 with one or two thousandths of its weight of charcoal, and 
 fuse the whole in a pot ; when cold, pulverise the litharge, 
 which will now be deprived of minium. 
 
 Pure carbon produces, with pure litharge, 34 times 
 its weight of lead, whilst hydrogen gives 103 '7 times its 
 weight of lead ; that is to say, a little more than three 
 times as much carbon. We can, therefore, from these 
 data, find the equivalent of any fuel, either in carbon or 
 hydrogen. 
 
 When a fuel contains volatile matters, the quantity 
 can be ascertained, as before pointed out, by ignition in a 
 close tube or crucible. If, further, we ascertain the pro- 
 portion of lead it gives with litharge, it is easy to calculate 
 the equivalent in carbon of the volatile matters, and, in 
 consequence, to ascertain its calorific value. 
 
 Supposing that a substance gives by distillation C parts 
 of coke, or carbon, having deducted the weight of the ash 
 and of volatile substances, and that it produces P parts of 
 lead with litharge. The quantity C of carbon would give 
 34 x C of lead ; the quantity of volatile matter would give 
 but P 34 x C ; it would be equivalent to * |^ of carbon : 
 whence it follows that the quantity of heat developed by 
 the charcoal, the volatile matter, and the unaltered com- 
 
DR. URK ON BERTH IERS METHOD. 157 
 
 bnstihle, will be to each other as the numbers 34x0,. 
 P-34xC, and P. 
 
 Dr. Ure* says, speaking of the above method of assay,. 
 ' On subjecting this theory to the touchstone of experi- 
 ment, I have found it to be entirely fallacious. Having 
 mixed very intimately 10 grains of recently calcined char- 
 coal with 1,000 parts of litharge, both in line powder, I 
 placed the mixture in a crucible, which was so carefully 
 covered as to be protected from all fuliginous fumes, and 
 exposed it to distinct ignition. 
 
 ' No less than 603 grains of lead were obtained, whereas, 
 by Berthier's rule, only 340 or 346*6 were possible. On 
 igniting a mixture of 10 grains of pulverised anthracite 
 with 500 grains of pure litharge previously fused and 
 pulverised, I obtained 380 grains of metallic lead. In a 
 second experiment, with the same anthracite and the same 
 litharge, I obtained 450 grains of lead ; and in a third,, 
 only 350 grains. It is therefore obvious that this method 
 of Berthier's is altogether nugatory for ascertaining the 
 quantity of carbon in coals, and is worse than useless 
 in judging of the calorific qualities of different kinds of 
 fuel.' 
 
 This discrepancy in the results obtained by Dr. Ure is 
 very perplexing, and does not at all accord with Berthier's 
 experience, as shown by his experiments, or by the author's 
 on the subject. The latter never had a difference of more 
 than 50 grains, and in general only two or three, which 
 latter result is satisfactory. The only precaution he found 
 necessary was to heat very gradually until the mixture 
 was fully fused, and then to increase the fire to bright 
 redness for a few minutes. 
 
 Further experiments have been made by the author 
 on this subject, and he has succeeded most perfectly in 
 estimating the value of a fuel. With the litharge of com- 
 merce, which contains much minium, the process is never 
 exact : results have been obtained differing as much as 40 
 or 50 grains when the litharge employed had not been 
 purified, and to purify it completely is a troublesome 
 
 * ' Supplement to the Dictionary of Arts, Mines, and Manufactures.' 
 
158 
 
 UEE'S CALORIMETER, 
 
 process. This difficulty may be completely obviated, 
 however, by substituting for litharge, white-lead, using 
 for each 10 grains of fuel 700 grains of white-lead, which 
 
 FIG. 66. 
 
 DIMENSIONS. 
 
 A B=3J in. ; diameter f in. 
 C D=2 in. 
 C B, socket for AB. 
 Diameter across D=4 in. 
 E F=6 in. 
 
 F G=5 in. ; diameter H G = l| in. 
 A B weighs 39 grammes. 
 The remainder of the apparatus, including 
 the stopcock, weighs 391 grammes. 
 
 SCALE OF 12 INCHES. 
 
 12 
 
 are well mixed with it, and 300 grains of pure white-lead 
 to cover the mixture. 
 
 When the whole is heated, the carbonate of lead de- 
 composes, forming pure lead oxide, which is then reduced, 
 as in the former case. By this process the results corre- 
 spond to 1 grain in the quantity of lead produced from a 
 
 
WRIGHT'S CALORIMETER. 159 
 
 given sample of fuel. Of course great care must be taken 
 that the white-lead is genuine. 
 
 Commercial samples are frequently adulterated with 
 lead sulphate and barium sulphate, lead oxychloride, zinc 
 oxide, &c. This is a serious drawback to this otherwise 
 excellent modification. 
 
 WRIGHT'S CALORIMETER. The instrument known as 
 Wright's calorimeter gives very accurate results, and is 
 the one most generally used now in experiments on the 
 heating power of fuel, in all but the most refined investi- 
 gations. It is shown in the accompanying figure (fig. 66). 
 
 The copper cylinder A B is filled with a mixture of 20 
 grains of the combustible, and 240 of the deflagrating com- 
 pound, which is composed of three parts of potassium 
 chlorate and one of potassium nitrate. A little piece of 
 cotton soaked in potassium chlorate is placed partly in the 
 mixture, the other end projecting above the top of the 
 cylinder ; this is ignited, quickly covered with the bell- 
 shaped part of the apparatus, and immersed in a measured 
 quantity of water. As constructed, the whole metallic 
 apparatus weighs 6,642'7 grains, and with this weight 
 290-1 grains of water are used. The temperature is re- 
 corded before and after making the experiment. During 
 the deflagration the stopcock is closed ; it is, however, 
 opened before taking the temperature the second time. A 
 tenth of the temperature that the water is raised by the 
 combustion is added for errors that are incidental to the 
 use of the instrument. 
 
 If the instrument is made of the weight above given, 
 the result is obtained by a very simple calculation. Each 
 Fahrenheit degree by which the temperature of the water 
 has been augmented corresponds to a pound of water 
 converted into steam. 
 
 EXAMPLE. 
 
 Fahr. 
 
 Temp, of water before making experiment = 56 
 after the combustion . = 65 
 
 ~9 + Ath 
 produced by the combustion = 9'1 
 
 One pound of the coal will convert into steam (maximum effect) 9'1 Ibs. of 
 water at 212 F. 
 
100 PYROMETKIC EXAMINATION OF FUEL. 
 
 5. ESTIMATION OF THE SPECIFIC HEATING POWER. This 
 represents the heat produced from a certain volume of 
 fuel. It may be ascertained by multiplying the absolute 
 heating power by the specific gravity. 
 
 6. ESTIMATION OF THE PYROMETRIC HEATING POWER. By 
 pyrometric heating power is meant the degree of tem- 
 perature which may be obtained by completely burning 
 the fuel. This heating power not only depends upon the 
 composition of the fuel, but chiefly on the time required 
 for its combustion, and this again depends on the looseness 
 and inflammability of the fuel. The absolute heating 
 power of hydrogen is greater than that of carbon, but with 
 regard to the pyrometric heating power it will be found 
 that the reverse is the case. 
 
 Carbon burned in contact with the air to carbonic 
 acid will produce a heat of 2,558 C. ; if burned to car- 
 bonic oxide it only produces 1,310 ; hydrogen burning to 
 water will produce a heat of 2,080. From this we learn 
 that fuel rich in carbon, such as anthracite, coal, and coke, 
 will produce a greater pyrometric effect than fuel rich in 
 hydrogen, as wood, &c. 
 
 Density is an essential quality of fuel required to pro- 
 duce great pyrometric effect. This is proved in the follow- 
 ing way. 
 
 When atmospheric air first acts on the carbon con- 
 tained in fuel, carbonic acid is formed, and the tempera- 
 ture rises to a certain degree, but on passing over glowing 
 coal, carbonic acid becomes converted into carbonic oxide, 
 and this causes a portion of the heat at first produced to 
 become latent. This conversion of carbonic acid into car- 
 bonic oxide is more easy and complete as the fuel used is 
 more inflammable ; and as v a greater quantity of heat is 
 thereby rendered latent, it follows that the heating power 
 of such a fuel is inferior. This accords with general ex- 
 perience ; for it is well known that coke is able to produce 
 a greater heat than charcoal. 
 
 Good methods for estimating pyrometric heating power 
 were given in the last chapter. 
 
 7. ESTIMATION OF THE VOLATILE PRODUCTS OF CARBONI- 
 
ASSAY OF FUEL. 161 
 
 SATION. The amount of volatile matter yielded on car- 
 bonising a fuel depends partly on the composition of the 
 fuel, and partly on the temperature employed. If a fuel 
 rich in oxygen and hydrogen is quickly heated, it will yield 
 the greatest amount of volatile products. These are partly 
 liquid (tar, naphtha, and acetic acid or ammoniacal water), 
 and partly gaseous (carbonic oxide, carbonic acid, and 
 light and heavy carburet ted hydrogen). The more oxy- 
 gen a fuel contains, the more carbonic acid and carbonic 
 oxide it will produce ; the more hydrogen it contains, the 
 more illuminating gas it yields. The applicability of a 
 sample of coal to the production of illuminating gas 
 depends on these conditions. 
 
 Coal distilled at a low temperature yields much tar and 
 comparatively little gas, and when a very high temperature 
 has been used, less tar and more gas is produced, but the 
 great heat will have reacted on the gas and injured its 
 illuminating qualities. If the coal contains pyrites, the 
 gas will contain sulphur compounds. The amount of 
 water produced is generally larger than that of the tar. 
 
 In order to estimate the amount of volatile matter given 
 off from any particular sample of coal, proceed in the fol- 
 lowing manner : Place a given weight, say 200 grains, of 
 the coal in an iron tube closed at one end, to the other 
 end of which adapt, by means of a cork, a glass or other 
 tube, which must be led into an inverted jar full of 
 water standing in the pneumatic trough. Eaise the tem- 
 perature very gradually to redness, and continue the heat 
 until no more gas is given off, then ascertain its quantity 
 in cubic inches, with due correction for temperature and 
 pressure. 
 
 8. EXAMINATION OF THE COKE OR CHARCOAL LEFT BEHIND ON 
 CARBONISATION. The amount of coke or charcoal yielded 
 by a sample of fuel is found by the last operation. This 
 residue is the amount of coke which that particular sample 
 of coal produces ; and its weight, divided by two, gives 
 the percentage of coke. 
 
 The process of coking, charring, or carbonising fuel, 
 whilst it drives off some of the valuable hydrocarbon con- 
 
 M 
 
162 EXAMINATION OF COKE OF FUEL. 
 
 stituents, also gets rid of all the aqueous elements. And 
 therefore the coke or charcoal which is left behind has 
 its value greatly increased when high temperatures are 
 required, although, from the absence of flame-yielding 
 constituents, it is much more difficult to ignite. 
 
 The degree of inflammability of coke or charcoal is 
 relatively the same as that of the raw fuel from which they 
 were produced. The more inflammable a fuel has been, 
 the more inflammable will be the coke or charcoal pro- 
 duced from it. 
 
 The temperature employed in the carbonisation, as has 
 been already explained, exerts great influence on the yield 
 of coke. 
 
 If the fuel contains iron pyrites, part of the sulphur 
 goes off in the volatile portion, but from one-fourth to 
 one-half is retained in the form of iron sulphide. 
 
 9. ESTIMATION OF THE AMOUNT OF ASH. In order to 
 ascertain the amount of ash : Fully ignite about 50 grains 
 of coal in a platinum capsule, allowing the air to have 
 free access all the time until nothing but ash is left. Its 
 amount may then be ascertained by weighing : good fuel 
 should contain little ash. It may vary from 1 to 10 per 
 cent., but if it exceeds 5 per cent, it becomes deleterious. 
 The chemical composition of the ash also influences the 
 quality of the fuel to some extent. 
 
 10. ESTIMATION OF THE AMOUNT OF SULPHUR. This is 
 an important operation in the assay, as a coal containing 
 sulphur cannot be employed for particular operations, and, 
 indeed, those which contain much sulphur ought only to 
 be used for the commonest purposes. This assay is most 
 important to ironmasters as well as to steamboat and other 
 companies, who consume fuel under steam boilers ; and 
 the coal they purchase should always be subjected to this 
 particular test, as sulphur has a corroding and destroying 
 action on iron and copper. Where sulphurous coals are 
 continually burnt under boilers, the metal of the latter 
 becomes deteriorated, and the boiler is rapidly rendered 
 useless. Sulphur exists in coal in the form of iron pyrites ; 
 this can generally be detected by its brassy colour. Some 
 
DETERMINATION OF SULPHUR IN FUEL. 1G:>> 
 
 coals and lignites also contain calcium sulphate, and in 
 rare cases barium sulphate. 
 
 The process for the estimation of the amount of 
 sulphur in coal is not difficult. 1 part of the coal is to be 
 finely pulverised, and then mixed with 7 or 8 parts of 
 potassium nitrate, 16 parts of common salt, and four parts 
 of potassium carbonate, all of which must be perfectly 
 pure ; the mixture is then placed in a platinum crucible 
 and gently heated. At a certain temperature the whole 
 ignites and burns quietly. The heat is then increased until 
 the mass is fused : the operation is finished when the mass 
 is white. It must, when cold, be dissolved in water, the 
 solution slightly acidulated by means of hydrochloric acid, 
 and barium chloride added to it as long as a white preci- 
 pitate forms. This precipitate is barium sulphate, which 
 must be collected on a filter, washed, dried, ignited, the 
 filter burnt away, and the remaining barium sulphate 
 weighed : every 116 parts of it indicate 16 of sulphur. 
 
 Dr. Price has drawn attention to a source of error 
 which lias hitherto escaped notice in the estimation of sul- 
 phur, where fusion of the substance with nitre is the pro- 
 cess employed. This author has found that unless great 
 care be taken to prevent the fused mass passing over to 
 the outside of the vessel, and so coming in contact with 
 the flame or products of combustion, an appreciable and, 
 in some cases, serious error will arise, owing to the sul- . 
 phuric acid produced from the sulphurous acid in the 
 flame a product of the oxidation of the sulphide of car- 
 bon contained in the gas combining with the potassium 
 of the fused salt. Several experiments have been made to 
 ascertain the amount of error that may be occasioned from 
 the above cause. In one instance the flame issuing from 
 a Bunsen's burner was made to strike against a little fused 
 nitre on the under side of a small platinum dish, when, in 
 three quarters of an hour, as much sulphuric acid was 
 obtained as is equivalent to 12 milligrammes of sulphur. 
 As a check on these experiments, nitre was fused by the 
 flame of the spirit lamp, when, as was to be anticipated, 
 not a trace of sulphuric acid could be detected upon the 
 
 IT 2 
 
164 DETERMINATION OF SULPHUR IN FUEL. 
 
 addition of a barium salt to the aqueous solution of this 
 fused mass, rendered acid by hydrochloric acid. In es- 
 timations of sulphur in coke or coal, great care should, 
 therefore, be taken to prevent any of the fused saline con- 
 tents of the crucible from getting on to the outside. In 
 fusing pig iron with nitre, a process recommended by some 
 for the estimation of the sulphur it contains, the mass, 
 especially if the iron be rich in manganese, invariably 
 creeps over to the outer wall of the crucible ; and it is, 
 therefore, impossible to obtain correct results when the 
 operation is conducted over the gas flame. The assayer 
 should for these reasons always employ a spirit flame in 
 preference to gas in sulphur estimations. 
 
 Mr. Teikichi Makamura recommends the following pro- 
 cedure : 3 or 4 parts of the mixed alkaline carbonates, or 
 of sodium carbonate, are intimately mixed with one part of 
 coal in very fine powder in a large platinum dish. The 
 mixture is heated at first very gently, a spirit lamp being 
 used instead of a Bunsen, to prevent possible absorption 
 of sulphur ; the heat is then raised slowly without attain- 
 ing that of visible redness, until the surface becomes 
 only faintly grey. No smoke or odorous gases should 
 escape during the whole of the oxidation. The tem- 
 perature is now raised to a faintly red heat for sixty 
 minutes, at the end of which time the mass is perfectly 
 white, or reddish if iron be present. The mass is not 
 to be stirred during the ignition. The residue is heated 
 with water, filtered, and the sulphates estimated in the 
 ordinary way. 
 
 11. EXAMINATION OF OTHER PECULIARITIES OF FUEL. 
 Besides the above-named examinations, the assayer should 
 notice the degree of inflammability of the fuel, and whether 
 any particular smell is evolved during combustion ; whether 
 the coal is good for coking purposes ; whether it burns 
 with a large smoky flame or a luminous flame ; whether it 
 burns quietly or with decrepitation ; and whether the ash 
 is dusty or fusible, and likely to accumulate and clog up 
 the grate-bars. 
 
 32. CALCULATION OF RESULTS. It may be sufficient here 
 
ASSAY OF FUEL. 165 
 
 to state that, beside the percentage composition of the 
 -coal, it is proper to reduce the composition to the com- 
 bustible =100, in order to obtain a comparative estimate 
 of the character of the fuel itself (in regard to the propor- 
 tion of bitumen and carbon), and of the amount and 
 quality of the impurities (ash and moisture). It has also 
 been shown that, for considerable areas of the coal-field, 
 the sum of the constituents on the scale of combustible = 
 100 is the proper calorific equivalent, and that the per- 
 centage of the combustible in the fuel gives a proper 
 estimate of its value. 
 
 ASSAY OP COAL BEFORE THE BLOWPIPE. The blowpipe 
 method is well adapted to the assaying of .coal. Not only 
 does the portableness of the apparatus make it very con- 
 venient for use away from home, wherever the balance 
 can be set up ; but its use at home is quite as satisfactory 
 on the score of exactness as the assay with the muffle or 
 retort, or large platinum crucible, and large balance. Mr. 
 B. S. Lyman gives the following directions for carrying 
 out this assay : 
 
 Besides the ordinary pieces of the blowpipe apparatus, 
 as made at Freiburg, all that needs to be made expressly 
 for the coal assay is a small covered platinum crucible of 
 the same size and shape as the clay crucibles of that appa- 
 ratus ; and there must be a little ring of German silver, 
 for the crucible to stand on, about three-eighths of an inch 
 across and half that in height. Such a crucible cover and 
 ring weigh about 40 grains more than the ordinary 
 metallic cup that rests on the pan of the balance ; the 
 crucible and ring without the cover weigh less than 30 
 grains more than the cup. If it be desired to estimate the 
 amount of hygroscopic moisture in the coal, a small dry- 
 ing bath must be made too ; but the hygroscopic water 
 in ordinarily well-dried coals (not brown coals) is of little 
 importance. 
 
 The size of the crucible allows the coking of 3 to 10 
 grains of coal, according to the dryness of the coal and 
 the extent of its swelling up when heated ; and as the 
 blowpipe balance weighs within l-l,000th of a grain, it 
 
16G ASSAY OF FUEL. 
 
 is easy to weigh within much less than l-10th of one per 
 cent, of the amount of coal assayed much nearer, in fact, 
 than the exactness of the coke assay in other respects. 
 On this point, indeed, the blowpipe assay is quite as good 
 as the assay with the larger balance, especially the muffle 
 assay, where the coal must be brushed into a clay recep- 
 tacle after weighing, and the coke or ashes brushed off 
 from it before weighing ; while here the crucible is 
 weighed each time without removal of its contents, and 
 without danger, therefore, of losing anything or adding 
 any dust. It may be objected that the smallness of the 
 amount of coal that can be assayed with the blowpipe 
 makes it a less trustworthy indicator of the general com- 
 position of the coal than a larger assay ; but the size of 
 the lumps or powder assayed may be made finer accord- 
 ingly, so that, when mixed up, an equally just sample of 
 the whole mass would be got for the small assay as for 
 the large. 
 
 Any one who has had a little experience, both in the 
 use of the blowpipe and in the ordinary muffle assay of 
 coal, will scarcely need any further teaching for the coal 
 assay with the blowpipe. For others it is worth while to 
 say that the coal may be assayed either in a fine powder 
 or in little lumps, and either with a slowly increasing or 
 with a quickly increasing heat. A quick heat will give 
 less coke by several per cents., but will often make a dry 
 coal cake together that would not cake with a slow heat. 
 The cover of the crucible should be left open a little, for 
 the easy escape of the gas, but covered enough to prevent 
 any flying off of solid material. The heat should increase 
 to redness, and as soon as the escaping gas stops burning 
 the heat should be stopped. As some coals part with their 
 gas more quickly than others, of course no definite time 
 can be fixed for heating all coals ; but the burning of the 
 gas is a good enough sign. Care should be taken not to 
 let the coke take up moisture from the air before weighing, 
 as it will quickly do if it has a chance. Of course, owing 
 to the different effects of quick or slow heating, a certain 
 uniformity of result, even with perfectly uniform samples 
 
ASSAY OF FUEL. 167 
 
 of coal, can only be got without error by practice and by 
 mechanical skill, by reproducing with nicety the same 
 conditions in successive assays. 
 
 After the coke has been weighed, it can be heated 
 again with very free access of air, say with the crucible 
 tilted to one side, and the cover off, until everything is 
 thoroughly burnt to ashes ; and these should be re-heated 
 until no change for the less is made in the weight. With 
 free-burning, soft (semi-bituminous) coals this burning to 
 ashes is very slow, so that it is very fatiguing or even im- 
 possible to carry it out with a blowpipe ; but in that case 
 the crucible may be heated over a Bunsen gas-burner or 
 an alcohol lamp, and left to glow for hour after hour. 
 The coking is far more conveniently done in the same way 
 than by blowing with the mouth. 
 
 As an illustration of the degree of accuracy which this 
 method may be expected to give, the author adduces a 
 pair of blowpipe assays, made five years ago, of some West 
 Virginia asphaltum, that seemed itself to be much more 
 uniform in composition than coal from different benches in 
 one bed is apt to be : 
 
 Volatile Matter Coke Ashes 
 
 No. 1 . . 47-29 per cent. 52-71 per cent. 1-65 per cent. 
 No. 2 . . 46-93 53-07 1-81 
 
 Mean . . 47-11 52'89 1'73 
 
 VALUATION OF COAL FOR THE PRODUCTION OF ILLUMINATING 
 GAS. Take 100 grains of the coal in small lumps, so that 
 they may be readily introduced into a rather wide com- 
 bustion tube. This is drawn out at its open end (after the 
 coal has been put in) so as to form a narrow tube, which 
 is to be bent at right angles ; this narrower open end is to 
 be placed in a wider glass tube, fitted tight into a cork 
 fastened into the neck of a somewhat wide-mouthed bottle 
 serving as tar vessel (hydraulic main of the gasworks). 
 The cork alluded to is perforated with another opening, 
 wherein is fixed a glass tube, bent at right angles for con- 
 veying the gas first through a chloride of .calcium tube, 
 next through. Liebig's potash bulbs containing a solution 
 of caustic potash, having lead oxide dissolved in it. Next 
 
168 ASSAY OP FUEL. 
 
 follows another tube, partly filled with dry caustic potash, 
 and partly with calcium chloride ; from this last tube a 
 gas-delivery tube leads to a graduated glass jar standing 
 over a pneumatic trough, and acting as gas-holder. Before 
 the ignition of the tube containing the coal is proceeded 
 with, all the portions of the apparatus are carefully weighed 
 and next joined by means of india-rubber tubing. After 
 the combustion is finished, which should be carefully con- 
 ducted, so as to prevent the bursting or blowing out of 
 the tube, the different pieces of the apparatus are discon- 
 nected and weighed again. The combustion-tube has to 
 be weighed with the coal after it has been drawn out at 
 its open end, and with the coke after the end of the com- 
 bustion when it is again cold ; and for that reason care is 
 required in managing it. We thus get the quantity of 
 tar, ammoniacal water, carbonic acid, and sulphuretted 
 hydrogen (as lead sulphide) ; and the gas is measured by 
 immersing the jar in water, causing it to be at the same 
 level inside and out. Empty the Liebig's bulbs into a 
 beaker, and separate the lead sulphide by filtration, wash 
 carefully, dry at 212 F., and weigh. From the lead sul- 
 phide the sulphuretted hydrogen present is calculated. 
 This process, devised by the late Dr. T. Eichardson, of 
 Newcastle-on-Tyne, was found by him to yield very trust- 
 worthy results, so as to be suitable for stating what 
 quantity of gas a ton of coal thus analysed would yield. 
 
109 
 
 CHAPTEE VI. 
 
 SEDUCING AND OXIDISING AGENTS FLUXES, ETC. 
 
 IN some assay operations in the dry way, bodies are heated 
 in suitable vessels per se ; but it is more often necessary 
 to add to the bodies submitted to assay other substances, 
 which are varied according to the nature of the change to 
 be effected. These substances may be divided into five 
 classes : I. reducing agents ; IE. oxidising agents ; III. 
 desulphurising agents ; IV. sulphurising agents ; and lastly, 
 V. fluxes properly so called. 
 
 I. EEDUCING AGENTS. 
 
 The substances belonging to this class have the power 
 of removing oxygen from those bodies with which it may 
 be combined. In assaying, the substance under examina- 
 tion is generally fully oxidised either naturally or artifi- 
 cially before reduction is required to be effected. The 
 most common reducing agents are as follows : 
 
 1. Hydrogen gas. 
 
 2. Carbon. 
 
 3. The fatty oils, tallow, pitch, and resins. 
 
 4. Sugar, starch, and gum. 
 
 5. Tartaric acid. 
 
 6. Oxalic acid. 
 
 7. Metallic iron and lead. 
 
 HYDROGEN GAS The most common method of prepar- 
 ing this gas consists in dissolving zinc in dilute sulphuric 
 acid. But as this plan gives the gas in the moist state, it 
 must be dried by being allowed to bubble through oil of 
 
170 EEDUCING AGENTS. 
 
 vitriol or by being passed through a bottle containing 
 fragments of dried calcium chloride before it is used for 
 assaying purposes. This gas will only be required in very 
 accurate assays, which are generally performed where 
 there are ample conveniences for generating pure hydro- 
 gen gas. The gas is colourless, invisible, and inodorous 
 when absolutely pure. It is a most powerful reducing 
 agent, and reduces a great number of metallic oxides at a 
 red or white heat, viz. the oxides of lead, bismuth, copper, 
 antimony, iron, cobalt, nickel, tungsten, molybdenum, and 
 uranium. When any metal is required in a state of abso- 
 lute purity, this is the only reducing agent admissible, as 
 others give the metal combined with a certain proportion 
 of carbon. 
 
 CARBON Found in large quantities in the mineral 
 kingdom, but generally combined with other bodies. In 
 a state of purity it constitutes the diamond. The diamond, 
 like all other species of carbon, is unacted on by the highest 
 possible temperature when in close vessels. It burns in 
 atmospheric air and oxygen gas, but requires for combus- 
 tion a higher temperature than ordinary charcoal. After 
 the diamond, the varieties of carbon found in nature or 
 artificially prepared are 
 
 1. BLACK-LEAD, or GRAPHITE. This is a mineral found 
 in beds in the primitive formations, principally in granite 
 and mica-schist. It is generally mixed with earthy sub- 
 stances, and rarely yields less than 10 per cent, of ash. 
 Before employing it for reduction purposes it should be 
 purified. Lowe * has given an excellent plan for effecting 
 this object. 
 
 ESTIMATION OP THE VALUE OP GRAPHITE. G. C. Witt- 
 stein f heats 15 grains of the sample to dull redness, and 
 takes the loss as water. The residue is ground up with 
 45 grains of a mixture of equal equivalents of potassium 
 and sodium carbonates, introduced into a platinum cru- 
 cible, covered with 15 grains of potash or soda, and slowly 
 heated to redness. The crust which is formed must be 
 
 * Polyt. Centr. 1855, p. 1404. f Dingler's Pqlyt. Journal, 216, 45. 
 
REDUCING AGENTS. 171 
 
 from time to time pushed down with a platinum wire. 
 After fusion for half an hour it is let cool, softened with 
 water, heated almost to a boil for a quarter of an hour, 
 filtered, washed well, and the liquid set aside. 
 
 The residue on the filter is dried, placed in a small 
 flask, the ashes of the filter added, and about 50 minims 
 of hydrochloric acid (sp. gr. 1-12) poured in. On the 
 mixture becoming slightly gelatinous a few more drops of 
 the acid are added, the digestion is continued for an hour, 
 the mixture diluted with water, filtered, and washed. Pure 
 graphitic carbon remains on the filter, and is dried, slightly 
 ignited, and weighed. The acid filtrate mixed with the 
 former one of an alkaline character is evaporated to 
 dryness, and in it silica, alumina, ferric oxide, &c., are 
 estimated in the usual manner. 
 
 2. ANTHRACITE. This is another species of fossil carbon 
 much resembling ordinary coal, but differing from it by 
 burning with neither smell, smoke, nor flame. 
 
 3. COKE. This is the residue of coal after all the 
 volatile matter is expelled. It is generally iron-black, 
 and has nearly a metallic lustre ; it is difficult to inflame, 
 and burns well only in small pieces, but gives a very in- 
 tense heat. Oven or furnace coke is preferable, as it is 
 harder, lasts longer, and is more economical in use. 
 
 4. WOOD CHARCOAL. This is obtained by burning the 
 woody part of plants with a limited supply of air, so as 
 to drive off all their volatile matters, and leave merely 
 their carbon. It is this kind that is generally employed 
 in assays. It ought to be chosen with care, well pulverised, 
 passed through a sieve, and preserved in well-stopped 
 vessels. Wood charcoal is never perfectly pure, generally 
 containing, besides ash, a proportion of hydrogen and 
 watery vapour ; these bodies are not generally prejudicial, 
 but in some experiments ash must not be present ; in that 
 case pure charcoal may be procured by heating sugar to 
 redness in a closed crucible. 
 
 The advantage of carbon as a reducing agent consists 
 in its great affinity for oxygen, which at a red heat sur- 
 passes that of most other substances. Charcoal by itself 
 
172 REDUCING AGENTS. 
 
 possesses two inconveniences : first, it lias the property of 
 combining with many metals ; and in the second place, it 
 is infusible, and cannot combine with vitreous substances. 
 The property it possesses of combining with iron, nickel, 
 cobalt, &c., is of no consequence to the assay er, for the 
 increase of weight it gives is not material, excepting under 
 the circumstances to be hereafter pointed out ; but its 
 infusibility and inability to combine with fluxes is a very 
 serious inconvenience ; for after the reduction, that portion 
 which has not been consumed remains disseminated with 
 the grains of metal in the fused slag, and prevents the 
 separation of all the metal, and the consequent formation 
 of a good button ; a large quantity of charcoal can thus 
 irreparably injure an assay. This inconvenience does not 
 happen, however, when an oxide is reduced by cementa- 
 tion in a lined crucible, but there are some cases in which 
 this mode of reduction is inadmissible. 
 
 Coke should never be used as a reducing agent in 
 assays when it is possible to avoid it. It often contains 
 a very large proportion of earthy and other extraneous 
 matters (more particularly sulphur, which is very in- 
 jurious). Coke is never so good as wood charcoal as a 
 reducing agent, because it burns more slowly. When it 
 is used, the temperature employed for an assay must be 
 much increased. 
 
 Coal is nearly always inconvenient, because it swells 
 by heat ; nevertheless, as it is not required in very large 
 quantities, it is sometimes employed, being finely powdered 
 and sifted previous to use. 
 
 THE FATTY OILS. The name oil is generally given to 
 those bodies that are fat and unctuous to the touch, more 
 Or less fluid, insoluble in water, and combustible. They 
 all become solid at various degrees of temperature. 
 There are some which, at the temperature of our climate, 
 have constantly a solid form, as butter, palm oil, cocoa- 
 nut oil, &c. 
 
 ; TALLOW is an animal product analogous to the fatty oils 
 in properties. 
 
 EESINS. The greater part of the resins are solid, but 
 
REDUCING AGENTS. 173 
 
 some are soft. They are brittle, with a vitreous and 
 shining fracture, and are often transparent. They are 
 very fusible, but cannot be raised to their boiling-point 
 without partial decomposition. 
 
 Although all the bodies just mentioned consume in 
 their combustion a large quantity of oxygen, they cannot 
 generally effect the total reduction of an oxide on account 
 of their volatility ; so that, before the temperature at 
 which the reduction takes place can be attained, the 
 greater part of the reducing agent has been expelled. 
 They generally act only by virtue of the small carbonaceous 
 residue produced by the action of heat ; so that their use 
 is very limited and uncertain. Whenever they are em- 
 ployed as reducing agents, without covering the substance, 
 a loss is experienced, on account of the bubbling and boil- 
 ing caused by their decomposition; this will always take 
 place unless the contents of the crucible be covered with 
 charcoal powder. Oils are very serviceable in the reduc- 
 tion of a large mass of oxide by cementation ; in this case, 
 after the oxide has been placed in the crucible, as much 
 oil is added as the oxide and the lining of the crucible will 
 soak up. Fat or resin is also used to prevent the oxida- 
 tion of the surface of a metallic bath (as in the fusion of 
 bar-lead samples), by coating the metal, and preventing 
 the action of the atmospheric oxygen. 
 
 SUGAR in its decomposition by heat leaves a much larger 
 proportion of carbon than the oils, fats, or resins ; so that 
 it would appear serviceable as a reducing agent. There 
 are some cases in which it may be used with advantage, 
 but it undergoes a great increase in volume when heated ; 
 so that losses in an assay may occur by the use of this 
 agent. To purify sugar from mineral ingredients it should 
 be re-crystallised from alcohol. It then may be used 
 as such, or after carbonisation. It yields about 14 per 
 cent, of charcoal : this is pure carbon, and leaves no resi- 
 due when burnt ; it is, therefore, preferable to wood char- 
 coal in cases where no foreign matter should be introduced 
 into the assay. 
 
 STARCH, well dried, and, better still, terrified, is em- 
 
174 KEDUCIXG AGENTS. 
 
 ployed with advantage as a reducing agent, and is better 
 than sugar, as it neither fuses, swells up, nor spirts, and in 
 many cases is even preferable to charcoal, because it is in 
 such a fine state of division that it can be more readily and 
 intimately mixed with the substance to be reduced. Wheat 
 and rye flour have nearly the same qualities as starch. 
 They are sometimes used. 
 
 GUM decrepitates slightly by heat, softens, agglomerates, 
 and boils, without spirting. The gums can be employed as 
 reducing agents under the same circumstances as sugar and 
 starch, but the two latter are preferable, because they con- 
 tain no earthy substances. 
 
 TARTARIC ACID is the reducing agent in cream of tartar, 
 or argol, of which so frequent use is made ; but the acid is 
 never employed by itself. When heated in close vessels 
 it fuses and decomposes, giving off combustible gases, leav- 
 ing a little charcoal. It burns when heated in contact 
 with air, giving rise to a peculiar and not unpleasant 
 odour. 
 
 OXALIC ACID fuses at a temperature of 208 without 
 decomposing, but when heated to 230 it is decomposed, 
 giving rise to carbonic acid, carbonic oxide, and a little 
 formic acid vapour ; and when heated strongly some por- 
 tions volatilise without decomposition : it does not leave a 
 carbonaceous residue. 
 
 The property which oxalic acid possesses of not leaving 
 a residue would render it remarkably valuable for the re- 
 duction of the metallic oxides in cases where the slightest 
 trace of carbon is to be avoided, if its reducing power were 
 greater ; but it decomposes at a low temperature, and in 
 burning absorbs but a small quantity of oxygen, especially 
 when it has not been dried ; so that even for the most 
 easily reducible oxides a large proportion must be em- 
 ployed. When it is combined with a base, as potash in 
 potassium binoxalate, its reducing power is much aug- 
 mented. 
 
 AMMONIUM OXALATE, when heated in close vessels, is 
 decomposed. Its reducing power is nearly double that 
 of oxalic acid. 
 
COMPARATIVE REDUCING POWER. 
 
 175 
 
 COMPARATIVE REDUCING POWER OP THE ABOVE AGENTS. In 
 order to give an idea of the comparative reducing power 
 of the agents just described, the result of some assays 
 made on them by Berthier, by means of litharge, are given 
 below. 
 
 By heating the same weight of each reducing agent 
 with an excess of litharge, buttons of leads were obtained 
 whose weights were proportional to the quantity of oxygen 
 absorbed, and by comparing them with each other the re- 
 ducing power of each flux is given ; by taking for unity 
 the weight of the reagent, calculation has proved that 
 1 part of pure carbon reduces from litharge 34-31 of 
 lead. The following are the results of Berthier's experi- 
 
 ments : 
 
 Hydrogen. 
 
 Pure carbon 
 
 Calcined wood charcoal 
 
 Amber resin 
 
 Ordinary wood charcoal 
 
 Animal oil 
 
 Tallow 
 
 Resin 
 
 Sugar 
 
 Torrified starch 
 
 Common starch 
 
 Gum Arabic 
 
 Tartaric acid 
 
 Ammonium oxalate . 
 
 Oxalic acid 
 
 104-00 
 
 34-31 
 
 31-81 
 
 30-00 
 
 28-00 
 
 17-40 
 
 15-20 
 
 14-50 
 
 14-50 
 
 13-00 
 
 11-50 
 
 11-00 
 
 6-00 
 
 1-70 
 
 90 
 
 It must be borne in mind that these numbers do not 
 represent the quantities of oxygen each reagent would ab- 
 sorb in complete combustion ; but that it only indicates 
 the quantity of metal produced by equal weights of the 
 reagents. 
 
 In assaying, however, it is rarely that these agents are 
 used by themselves ; they are generally mixed with a flux 
 properly so called, which will be more particularly de- 
 scribed under the head of Fluxes. 
 
 METALLIC IRON removes oxygen from the oxides of lead, 
 bismuth, copper, &c., but is rarely added for that especial 
 purpose ; and when it does produce this effect it is gener- 
 ally secondary, because it previously existed in the matter 
 subjected to assay, or was added for some other purpose. 
 
176 OXIDISING AGENTS. 
 
 METALLIC LEAD reduces but a very small number of 
 oxides, but it reduces many to the minimum of oxidation ; 
 it also decomposes some sulphates and arseniates. 
 
 II. OXIDISING AGENTS. 
 
 The oxidising agents in general use are as follows : - 
 
 1. Oxygen, atmospheric or combined. 
 
 2. Litharge and ceruse. 
 
 3. Lead silicates and borates. 
 
 4. Potassium nitrate. 
 
 5. Lead nitrate. 
 
 6. Manganese peroxide. 
 
 7. Copper oxide. 
 
 8. Iron peroxide. 
 
 9. The caustic alkalies. 
 
 10. The alkaline carbonates. 
 
 11. Lead, copper, and iron sulphates.- 
 
 12. Sodium sulphate. 
 
 LITHARGE is a fused protoxide of lead, and is generally 
 obtained from the silver-lead works. When melted, it 
 oxidises nearly all the metals, except mercury, silver, gold, 
 palladium, platinum, &c., and generally forms very fusible 
 compounds with the oxides. These two properties cause 
 it to be a very valuable agent in separating silver and 
 gold from all the substances with which they may be 
 mixed. 
 
 Litharge is occasionally mixed with a little of the red 
 oxide of lead ; the presence of this in large quantities be- 
 comes injurious, as it has the property of oxidising silver. 
 Ordinary litharge can be easily freed from this oxide by 
 fusing it and pouring it- into a cold ingot mould, then 
 pulverising, and carefully keeping it from contact with air, 
 as it readily absorbs oxygen, and if it be allowed to cool in 
 the atmosphere it will nearly all be converted into the red 
 oxide. 
 
 CEKUSE, or WHITE-LEAD, is a carbonate of lead protoxide. 
 
OXIDISING FLUXES. 177 
 
 As it does not contain the slightest traces of red oxide, it 
 may be used where the presence of that substance may be 
 inconvenient ; but it is troublesome to use, as it is much 
 less dense than litharge ; large vessels must be employed 
 in consequence ; besides, it generally contains a small 
 quantity of lead acetate or subacetate, and sometimes 
 metallic lead separates from it on ignition, which is, in 
 some cases, disastrous to the result of an experiment. 
 When ceruse is employed, a certain quantity must be fused, 
 to ascertain if any metallic lead be produced ; and, on the 
 other hand, it must be examined to ascertain if it be adul- 
 terated with barium sulphate. When it is pure it dissolves 
 completely in acetic or nitric acid. 
 
 LEAD SILICATES AND BORATES behave as litharge, but they 
 oxidise less rapidly. 
 
 They may be prepared by fusing together 1 part of 
 silica or boracic acid with 1 part of litharge. The borates 
 are more fusible than the silicates, but their use is attended 
 with inconvenience, as they swell very much in fusing. 
 
 POTASSIUM AND SODIUM NITRATES fuse at a temperature 
 below redness without alteration, but when heated more 
 strongly they give up oxygen. The action of these salts, 
 when fused, is very energetic, because they have a great 
 tendency to decompose, and because they contain a large 
 quantity of oxygen. They are used as oxidising agents in 
 the purification of the noble metals, and for ' preparing 
 some fluxes. They ought always to be employed in a 
 state of purity. 
 
 Saltpetre often contains impurities. On this account 
 an estimation of the real amount of potassium nitrate 
 often becomes necessary, not only in cases where saltpetre 
 is to be used for docimetric purposes, but also when used 
 in certain technical operations, viz. the manufacture of 
 gunpowder, enamel, &c. 
 
 If saltpetre is very impure, it may easily be purified 
 by recrystallisation to such a degree that it will only con- 
 tain 2 to 3 per cent, foreign substances (chiefly sodium 
 chloride). 
 
 An exact assay of saltpetre is most difficult, and the 
 
178 IMPURITIES IN SALTPETRE. 
 
 different modes in use are not quite accurate, on ac- 
 count of the nitrate of soda generally present as an 
 impurity in saltpetre, and not easy to estimate by means 
 of reagents. 
 
 Manufacturers often intentionally mix the raw salt- 
 petre with soda-saltpetre, and it is also often manufac- 
 tured from a mixture of soda-saltpetre and potassium 
 carbonate. 
 
 ASSAY OF SALTPETEES. The following are the different 
 modes of assaying saltpetre. 
 
 a. To estimate the nitric acid directly, 150 grains of 
 the well-ground sample are mixed with six times its weight 
 of well-powdered silica, placed in a platinum capsule and 
 dried in the water-bath till there is no further loss of 
 weight. The temperature is then raised to dull redness 
 for about half an hour, by which time the whole of the 
 nitric acid is expelled. The difference in weight gives the 
 nitric acid. 
 
 Sodium nitrate may be treated in the same manner. 
 
 The most usual, though not the most satisfactory, pro- 
 cess for the assay of potassium and sodium nitrates is to 
 estimate the total impurities, viz. water, insoluble matter, 
 alkaline chlorides and sulphates, and the residue being 
 taken as pure nitrate. 
 
 b. Gay-Lussac's mode of assaying saltpetre consists in 
 converting the potassium nitrate into potassium carbonate, 
 and in estimating its amount volu metrically by means of 
 standard sulphuric acid. 2-639 grains of saltpetre are 
 mixed with 1 grain of ignited pine-root, and 12 grains of 
 ignited and finely pulverised sodium chloride (the latter 
 is added in order to moderate the combustion), and this 
 mixture is heated in a platinum crucible. After cooling, 
 the mass is extracted by water, and either a standard solu- 
 tion of sulphuric acid or oxalic acid is added to the solu- 
 tion. The sulphuric acid is prepared by mixing 70 grains 
 of sulphuric acid, sp. r gr. 1*84, with 600 grains of water, 
 and to this mixture so much water is added again that 
 100 measures of it will saturate 6-487 grains of potassium 
 carbonate. The number of measures used for saturation 
 
ASSAY OF SALTPETRE. 179 
 
 will then indicate directly the percentage of potassium 
 carbonate. 
 
 The following foreign substances in raw saltpetre 
 should be estimated. 
 
 Water. 150 to 300 grains of air-dried, '.'finely pul- 
 verised saltpetre are heated in a porcelain crucible to 
 120 C. 5 and the resulting loss is calculated as water. 
 
 Mechanically mixed Impurities. The substance ob- 
 tained in the former assay is dissolved in hot water, and 
 filtered through a dried and weighed filter. The residue 
 is well washed with hot water, dried on the filter at 
 250 F., and weighed. On deducting the weight of the 
 filter, there will be left the weight of the mechanically 
 mixed impurities (alumina, silica, calcium carbonate, iron 
 peroxide, &c.), which usually amount to 2 to 5 per cent. 
 
 Lime and Magnesia. These substances are precipitated 
 as carbonates by sodium carbonate, in the former filtered 
 solution raised to the boiling-point ; the carbonates are 
 then dissolved in hydrochloric acid, and neutralised with 
 ammonia. The lime can be precipitated by oxalic acid, 
 and filtered off; the magnesia which remains in solution 
 may then be precipitated by sodium phosphate. 
 
 The amount of lime in East Indian raw saltpetre which 
 has been once crystallised varies between 0*21 and 0-26 
 per cent., the amount of magnesia between 0-26 and 0'28 
 per cent. 
 
 Chlorine. Thirty or 40 grains of raw saltpetre are dis- 
 solved in about an ounce of pure warm water in a flask 
 furnished with a tight-fitting stopper, and the amount of 
 chlorine is estimated by a standard solution of nitrate of 
 silver. The solution, after being warmed and acidulated 
 with nitric acid, is mixed gradually with the solution of 
 silver ; after each addition of the latter it is to be shaken 
 and then allowed to rest. (For particulars see the chapter 
 on the assay of silver in the wet way.) 
 
 The amount of chlorine found by this assay is cal- 
 culated as being derived from f potassium chloride and 
 ^ sodium chloride, so that 1 part of chlorine corresponds 
 to 1-927 part of metal (1-285 potassium, 0-642 sodium). 
 
 N 2 
 
180 ASSAY OF SALTPETRE. 
 
 Experience has proved that East Indian saltpetre contains 
 potassium and sodium chlorides in these proportions. 
 
 Sulphuric Acid. 100 to 120 grains of raw saltpetre 
 are dissolved in six ounces of water, and from this solu- 
 tion, heated to the boiling-point, the sulphuric acid is 
 precipitated by means of a solution of barium nitrate. 
 The precipitated barium sulphate is filtered off, washed , 
 ignited, and weighed. The amount of sulphuric acid in 
 East Indian raw saltpetre varies between 0-05 and (Ml 
 per cent. 
 
 Sodium Nitrate. This estimation is most difficult, arid 
 the following modes are recommended. 
 
 a. Longchamps's mode is based upon the decomposition 
 of soda-saltpetre by potassium chloride, producing sodium 
 chloride and potassium nitrate. The saltpetre is mixed 
 with potassium chloride, and the solution evaporated 
 down. By this operation, first sodium chloride and after- 
 wards saltpetre become separated. The latter is washed, 
 dried at 150 C., and weighed. Werther has recommended 
 a similar mode. 
 
 3. If the saltpetre does not contain certain oxides, 
 such as alumina, lime, &c. (or if, previously present, they 
 have been precipitated), a solution of potassium antimo- 
 niate will precipitate the soda contained in the saltpetre 
 solution. The precipitate consists of sodium antimoniate, 
 100 parts of which contain 84*39 antimonious acid, and 
 15-61 soda. 
 
 7. The presence of soda is also to be ascertained by 
 washing saltpetre with a saturated solution of pure potash- 
 saltpetre. This saturated solution will then contain a pro- 
 portionally large amount of sodium nitrate. If a small 
 quantity of the solution is made to crystallise upon a 
 watch-glass, soda-saltpetre, showing a rhombohedricform, 
 may be detected by means of a microscope, while potash- 
 saltpetre crystallises in prisms, and sodium and potassium 
 chlorides in cubes arranged in the form of steps. 
 
 LEAD NITRATE acts in a similar way to the two last- 
 mentioned salts. It is prepared by dissolving litharge in 
 nitric acid, and crystallising the solution. 
 
DESULPHURISING REAGENTS. 181 
 
 MANGANESE PEROXIDE is easily reduced to the state of 
 protoxide by many metals, and is a very powerful oxi- 
 dising agent ; but is rarely employed, because its com- 
 pounds are very infusible. It is employed occasionally 
 in the purification of gold and silver. 
 
 COPPER OXIDE is not much employed as a Hux, but is 
 often contained in substances submitted to assay ; it then 
 acts as an oxidising agent. A great number of metals, 
 ven silver, reduce it to the minimum of oxidation ; and 
 other metals, as iron, for instance, totally reduce it. 
 
 IRON PEROXIDE. This, like copper oxide, sometimes acts 
 incidentally as an oxidising agent. 
 
 THE CAUSTIC ALKALIES, POTASH AND SODA, fuse below a 
 red heat, and volatilise sensibly at a higher temperature. 
 Charcoal, at a high temperature, decomposes the water 
 combined with the hydrates of potash and soda, convert- 
 ing them into carbonates, but an excess at a white heat 
 decomposes the carbonate, and potassium or sodium is 
 the product. 
 
 POTASSIUM AND SODIUM CARBONATES are very much em- 
 ployed as agents in the assay by the dry way. They 
 have the power of oxidising many metals, as iron, zinc, and 
 tin, by the action of the carbonic acid they contain ; part of 
 it being decomposed, with the formation of carbonic oxide. 
 
 LEAD, COPPER, AND IRON SULPHATES. These three salts 
 at a high temperature oxidise the greater number of the 
 metals, even silver, the sulphuric acid giving off oxygen 
 and sulphurous acid. They are used in the assay of gold. 
 
 SODIUM SULPHATE is not used by itself as a reagent, but 
 is a product in many operations ; it is either formed in 
 the course of an assay, or is contained as an impurity in 
 some of the bodies used. 
 
 III. DESULPHURISING EEAGENTS. 
 
 1. The oxygen of the atmosphere. 
 
 2. Charcoal. 
 
 3. Metallic iron. 
 
 4. Litharge. 
 
 5. The caustic alkalies. 
 
182 LITHARGE. 
 
 6. The alkaline carbonates. 
 
 7. Nitre. 
 
 8. Lead nitrate. 
 
 9. Lead sulphate. 
 
 1. THE OXYGEN OF THE ATMOSPHERE acts as a desulphuris- 
 ing agent in roasting, combining with the sulphur present,, 
 forming sulphurous acid or sulphuric acid, sometimes 
 both. 
 
 2. CHARCOAL decomposes many sulphides by taking 
 their sulphur to form sulphide of carbon. It acts in this 
 manner with the mercury, antimony, and zinc sulphides. 
 It is only employed as an auxiliary to the desulphurising 
 power of the alkalies and their carbonates. 
 
 3. IRON separates sulphur from lead, silver, mercury, 
 bismuth, zinc, antimony, and tin, but only partially decom- 
 poses copper sulphide. It is generally used in the state 
 of filings, or nails ; the latter are preferable, and ought to 
 be kept free from rust. Oxide of iron may be used if it 
 be mixed with the requisite quantity of charcoal to reduce 
 it. Cast iron must not be employed, as it has very little 
 affinity for sulphur. 
 
 4. LITHARGE exercises a very energetic action on sul- 
 phides, even at a low temperature. If it be employed 
 in sufficient proportion, the sulphide acted on is wholly 
 decomposed. The sulphur is often disengaged as sulphur- 
 ous acid, and the metal remains alloyed with the lead 
 proceeding from the reduction of a portion of the litharge, 
 or combines as oxide with that portion of the litharge 
 which is not reduced. The quantity of litharge requisite 
 for the decomposition of a sulphide is considerable, and 
 varies according to its nature ; some sulphides require 
 34 times their weight. When less than the requisite 
 quantity is used, only a portion of the sulphide is decom- 
 posed, and a corresponding quantity only of lead reduced, 
 whilst the remainder of the sulphide forms, with the 
 litharge and the metallic oxide which can be produced, 
 a compound belonging to the class of oxysulphides, which 
 is generally very fusible. 
 
ACTION OF LITHARGE ON SULPHIDES. 183 
 
 When the sulphides have a very strong base, as an 
 alkali or alkaline earth, no sulphurous acid is given off by 
 the action of litharge, but all the sulphur is converted into 
 sulphuric acid. 
 
 Litharge is a very valuable reagent, and its use is 
 nearly exclusively confined to the assay of sulphides 
 containing the noble metals, as these metals are thus 
 obtained as alloys of lead, which are afterwards assayed 
 by cupellation. 
 
 The following is an account of the behaviour of this 
 reagent with the ordinary sulphides. 
 
 Manganese Sulphide requires at least six times its 
 weight of litharge to produce a fusible compound, and 
 thirty times its weight to desulphurise it completely. The 
 sulphur and metal oxidise simultaneously, and a manganese 
 protoxide is formed, which partly peroxidises, taking a 
 brownish tint in contact with the atmosphere. Berthier 
 assayed the four following mixtures : 
 
 Manganese sulphide . . 5 5 5 5 
 
 Litharge . ... 20 30 100 150 
 
 The first produced an infusible, greyish-black, scori- 
 forin mass, in which small plates, having the look of 
 galena, could be discovered. It was composed of the 
 sulphides and oxides of manganese and lead. Much 
 sulphurous acid was given off during the operation. 
 
 The second fused to a soft paste, and gave 17' 5 of 
 lead, and a compact, vitreous, opaque slag, of a very deep 
 brown colour. The slag contained about half its weight 
 of manganese sulphide. 
 
 The third fused readily, and produced 31-5 of ductile 
 lead, and a transparent, vitreous slag, of a deep hyacinth- 
 red. 
 
 The fourth produced 33'T of lead, exceedingly ductile, 
 and the desulphurisation was complete. 
 
 Iron Sulphide. -Thirty; parts of litharge are sufficient 
 to scorify iron protosulphide ; the metal is converted into 
 the protoxide. 
 
184 ACTION OF LITHARGE ON SULPHIDES. 
 
 The four following mixtures 
 
 Iron protosulphide ... 10 10 10 10 
 
 Litharge 60 125 250 300 
 
 gave, the first a pasty, scoriform mass, colour metallic 
 grey, and very magnetic. It was composed of the sul- 
 phides and protoxides of iron and lead. 
 
 The second gave a very fluid metallic black slag, very 
 magnetic, opaque, and possessing great lustre, and 36 of 
 lead. 
 
 The third gave a compact vitreous transparent slag of 
 a fine resin-red, and 67 of lead. 
 
 The last yielded a similar slag to the former, but con- 
 taining no sulphur, and 70 of lead. 
 
 Native iron pyrites was treated with the following 
 proportions of litharge : 
 
 Iron pyrites . . 10 10 10 10 10 10 
 Litharge. . . 60 125 200 300 400 500 
 
 The mixtures fused very readily with an abundant 
 disengagement of sulphurous acid. 
 
 The first produced only a metallic button, divisible 
 into two parts : the lower was the larger, and was a lead 
 subsulphide ; the other looked like compact galena, but 
 was magnetic ; it was composed essentially of the sulphides 
 of iron and lead, but probably contained a small quantity 
 of their oxides. 
 
 The second and third gave black vitreous opaque 
 slags which stained the crucibles brown, together with 
 lead, having a granular fracture and a deep grey colour ; 
 the first button weighed 35, and the second 40. Both 
 samples of lead were contaminated with a small quantity 
 of slag, and contained from y^Vo ^ Tiro f sulphur, and 
 a small quantity of iron. 
 
 The slags from the last three mixtures were vitreous, 
 transparent, and of a fine resin-red colour : the buttons of 
 lead equalled 45-4, 54*8, and 86 parts. A much larger 
 proportion of litharge does not produce more than 86 of 
 lead, proving that 50 parts of litharge completely effect 
 the desulphurisation of iron pyrites. 
 
ACTION OP LITHARGE OJS T SULPHIDES. 185 
 
 Copper Sulphide. The following mixtures of sulphide 
 of copper and litharge 
 
 Copper sulphide . .10 10 10 10 10 
 
 Litharge .... 20 30 50 100 250 
 
 fuse very readily, giving off an abundance of sulphurous 
 acid. 
 
 The slags formed were compact, vitreous, opaque, 
 or translucid, and more or less bright red. The copper 
 which they contained was at the minimum of oxidation. 
 
 The first three mixtures gave metallic buttons com- 
 posed of uncombined lead and sulphide of copper. 
 
 The fourth gave 28 of lead, with a little adhering 
 sulphide of copper. 
 
 The fifth gave 38*5 of pure ductile lead, the exact 
 quantity that ought to be reduced from litharge by the 
 transformation of the above quantity of copper sulphide 
 into suboxide and sulphurous acid. 
 
 Copper sulphide does not combine with litharge ; this 
 is an exception to the general rule. It requires about 
 twenty-five times its weight of litharge to decompose it 
 completely. When litharge is combined with a certain 
 quantity of copper protoxide, it -has no action on the 
 sulphide of that metal. 
 
 The desulphurisation of copper pyrites requires about 
 30 parts of litharge. 
 
 Copper pyrites .... 10 10 10 10 
 
 Litharge ..... 50 100 200 300 
 
 were fused together. 
 
 In the first assay the fusion was accompanied with 
 much ebullition, and the mass remained pasty : 6 parts of 
 ductile lead were produced, and a matte similar to galena, 
 but deep grey, with small facets, and a brownish-black 
 vitreous slag. 
 
 In the second, much ebullition and swelling up took 
 place : 35 of lead, 45 of matte, and a deep brown vitreous 
 slag were produced. 
 
 In the third assay, 49 of lead was the result. It was 
 covered by a thin layer of matte, and a very shining, dee]) 
 brown, vitreous, translucid slag. 
 
18G ACTION OF LITHARGE ON SULPHIDES. 
 
 The last mixture fused readily, almost without ebul- 
 lition, and gave 72 of lead, and a compact shining slag, of 
 a bright grey, and without the least trace of matte ; the 
 desulphurisation was complete. 
 
 Antimony Sulphide has a great tendency to combine 
 with litharge, and it must be heated with at least 25 parts 
 to effect its desulphurisation. By mixing these two sub- 
 stances in the following proportions 
 
 Antimony sulphide .10 10 10 10 10 
 
 Litharge . . .38 60 100 140 250 
 
 the first three mixtures afforded very fluid slags, compact, 
 deep black, and slightly metallic, and. buttons of ductile 
 lead, weighing 2, 9, and 26 parts. These slags resemble 
 the black litharge produced at the commencement of a 
 cupellation. 
 
 The fourth mixture gave a transparent compact slag, 
 vitreous and shining, having a splendid hyacinth-red 
 colour, and 50 parts of lead. 
 
 The last produced 57 of lead, proving the desulphuri- 
 sation to be complete. The antimony, in this case, exists 
 as protoxide in the slag. 
 
 M. Fournet has observed that antimony sulphide has 
 the property of carrying copper sulphide, and even silver 
 sulphide, into the compounds formed with litharge. In 
 one of the experiments which he made, a double sulphide, 
 composed of equal parts of silver sulphide and antimony 
 sulphide, was fused with three times its weight of litharge, 
 and gave, first, a button of lead mixed with silver ; 
 secondly, a matte like galena ; and thirdly, a black slag. 
 This slag was analysed, and found to contain from 8 to 9 
 per cent, of silver. 
 
 It is probable that all the sulphides, having a strong 
 tendency to combine with lead oxide, have, like antimony 
 sulphide, the property of determining the scorification 
 of a certain quantity of silver sulphide ; like all the sul- 
 phides, which, in a state of purity, are completely decom- 
 posed by lead oxide. 
 
 Zinc Sulphide must be fused with twenty-five times its 
 
ACTION OF LITHARGE OX SULPHIDES. 187 
 
 weight of litharge to be decomposed. The following mix- 
 tures were heated together : 
 
 Blende .... 24-08 12-08 10 10 
 
 Litharge ... . 55-78 83-68 100 250 
 
 However strongly the first mixture was heated, it 
 always remained pasty ; 29*2 of a greyish-black lead were 
 produced, which contained -018 of sulphur and -008 of 
 zinc. The button was covered by a metallic-looking black 
 substance, intermediate between a matte and a slag ; it 
 was composed of zinc and lead sulphides and oxides. 
 
 The second mixture gave 35 -5 of lead and a fluid slag, 
 which was compact, opaque, and black. 
 
 The third gave 43 of lead, and a deep grey slag. 
 
 The last produced 65 of pure lead, and a vitreous slag, 
 of an olive colour, and translucid on the edges. 
 
 Lead Sulphide. Galena and litharge, at a heat just 
 sufficient to fuse them, combine and form an oxysulphide; 
 but if the temperature be increased, the two bodies react 
 on each other, and are mutually decomposed. If 2,789 
 parts of litharge be employed to 1,496 of lead, or 1,865 of 
 litharge to 1,000 of galena, nothing but pure lead is 
 obtained. If more litharge be employed, a portion is not 
 decomposed, and covers the lead. If less be employed, 
 the galena is not completely decomposed, and the lead is 
 covered by a matte of subsulphide. 
 
 But when litharge is combined with a certain prp- 
 portion of sulphides or metallic oxides, it completely loses 
 its oxidising power on galena, even at a white heat ; so 
 that it can be combined with this substance as with the 
 other sulphides, without effecting its total decomposition. 
 
 5, 6. CAUSTIC ALKALIES AND THEIR CARBONATES. All the 
 sulphides are decomposed by caustic alkalies and their 
 caroonates; but in the latter case carbonaceous matter 
 must be present. In the absence of charcoal there are 
 some sulphides, as of copper, on which they have no 
 action. In these decompositions alkaline sulphides are 
 formed, and combine with and retain a certain quantity 
 of the sulphide submitted to experiment. The proportion 
 
188 CAUSTIC ALKALIES AND THEIR CARBONATES. 
 
 of the sulphide which remains in combination with the 
 alkaline sulphides depends on many circumstances. It is 
 always less when a large proportion of alkali or carbonate 
 has been employed, as it is also when a high degree of 
 temperature has been employed ; and the presence of 
 charcoal always much diminishes the proportion. When 
 the metal of a sulphide is very volatile, as mercury or 
 zinc, the decomposition may be perfect- 
 Potash, as it is sold in commerce, generally contains 
 foreign substances, viz. silica, peroxide of iron, potassium 
 sulphate, chloride, phosphate, and silicate, soda-salts, &c., 
 and also water. 
 
 A partial purification of the potash may be effected by 
 dissolving it in water, which will not dissolve some of the 
 above-named foreign substances. 
 
 Soda also is never free from foreign substances. 
 Ammonium carbonate is used for decomposing metallic 
 sulphates which are formed during the roasting process of 
 several sulphur minerals. Ammonium sulphate is then 
 formed, which is volatile when slightly heated. 
 
 7. NITRE, SALTPETRE, OR POTASSIUM NITRATE has a very 
 powerful action on the sulphides ; in fact, if not modified 
 by the addition of some inert substance, as an alkaline 
 carbonate or sulphate, explosion may take place, and a 
 portion of the contents of the crucible be thrown out. 
 Where an excess of nitre is used, all the sulphur is con- 
 verted into sulphuric acid, and every metal but gold and 
 silver is oxidised. When only the exact quantity of nitre 
 is employed that is to say, just as much as is sufficient 
 to burn all the sulphur in the sulphides of those metals 
 which are not very oxidisable, as those of copper, silver, 
 and lead, the metal is obtained in a state of purity, and 
 the whole of the sulphur converted into sulphuric acid ; 
 but with the sulphides of the very oxidisable metals the 
 oxygen of the nitre is divided between the sulphur and 
 the metal. 
 
 8. LEAD NITRATE possesses the combined properties of 
 nitre and litharge. It is not much used. 
 
 9. LEAD SULPHATE is not used as a reagent, but is often 
 
SULPHURISING REAGENTS. 189 
 
 formed in the assay of lead ores. It decomposes lead 
 sulphide by burning the sulphur. It acts on many other 
 sulphides in a similar manner. 
 
 IV. SULPHURISING REAGENTS. 
 
 1. Sulphur. 
 
 2. Cinnabar, or mercury sulphide. 
 
 3. Galena. 
 
 4. Antimony sulphide. 
 
 5. Iron pyrites. 
 
 6. The alkaline persulphides. 
 
 1. SULPHUR fuses at 226, and at 284 is very liquid. 
 It has very powerful affinities, and combines with the 
 greater number of the metals. That kind generally known 
 as flowers of sulphur ought to be employed ; and before 
 use, the presence or absence of earthy matters should be 
 ascertained by exposing it to a dull red heat in a crucible. 
 The sulphur will go off, and the earthy impurities will be 
 left behind. 
 
 Sulphur is principally used in the preparation of the 
 alkaline sulphides and in the assay of some of the noble 
 metals. 
 
 2. CINNABAR is decomposed by many of the metals, and 
 is a better sulphurising agent than sulphur itself, as it is 
 less volatile. 
 
 3. GALENA. Many metals, as iron, copper, &c., separate 
 sulphur from lead, while some others, as silver, gold, &c., 
 do not ; so that if galena be heated with an alloy of various 
 metals, some of which decompose it, and some do not, the 
 former are transformed into sulphides, and the latter com- 
 bine with the metallic lead which is produced. It is often 
 employed for this purpose. It is a common ore, and 
 readily procured. 
 
 The samples employed must contain no antimony sul- 
 phide, and alHhe matrix must be carefully separated by 
 sifting and washing. 
 
 4. ANTIMONY SULPHIDE yields its sulphur to many of 
 the metals, but it is only used in the separation of gold 
 
190 SULPHURISING REAGENTS. 
 
 from some alloys. In this operation the sulphur combines 
 with the alloyed metals, and the antimony with the gold, 
 for which it has much affinity. ; 
 
 5. IRON PYRITES is a persulphide which loses half its 
 sulphur at a white heat. It is much employed in metal- 
 lurgical operations, but not in assaying. 
 
 6. ALKALINE PERSULPHIDES can support a tolerably 
 elevated temperature without losing sulphur, but they 
 have a great tendency to do so, and to this is due their 
 sulphurising power. By their means almost every metal 
 can be made to combine with sulphur. When an alkaline 
 persulphide is heated with a metal, or an oxide of a metal 
 mixed with charcoal, a fused compound, a mixture of 
 the sulphide of the metal and an alkaline sulphide, is 
 obtained. 
 
 When they are in combination they are held together 
 by very feeble affinities, and their decomposition is gener- 
 ally eifected by the mere action ; of water, which dissolves 
 the alkaline sulphide and leaves the other perfectly pure. 
 But with gold, molybdenum, tungsten, antimony, &c., the 
 compound is stable and soluble in water ; and it is from 
 this fact that the alkaline sulphides are sometimes em- 
 ployed in the assay of auriferous substances. 
 
 In order to effect a sulphurisation by means of the 
 alkaline sulphides, it is much better to use equivalent 
 mixtures of sulphur and alkaline carbonates than to pre- 
 pare them beforehand. To obtain potassium persulghide, 
 46 parts of potassium carbonate, and 54 of flowers of 
 sulphur, must be fused together ; and for sodium persul- 
 phide, 40 parts of dry sodium carbonate must be heated 
 with 60 parts of sulphur. > 
 
 When the mixture is fused in a plain crucible, potas- 
 sium sulphate or sodium sulphate is formed, because part 
 of the alkali gives up its oxygen to a portion of the 
 sulphur, which becomes sulphuric acid ; but when char- 
 coal-lined crucibles are used, the carbon combines with 
 the oxygen of the alkali, and no sulphate is produced. 
 
FLUXES. 191 
 
 V. FLUXES. 
 
 Fluxes are used in the following cases : 
 
 1st. To cause the fusion of a body either difficultly 
 fusible or infusible by itself. 
 
 2ndly. To fuse foreign substances mixed with a metal, 
 in order to allow the latter to separate by its difference of 
 specific gravity. 
 
 ordly. To destroy a compound into which an oxide 
 enters, and which prevents the oxide being reduced by 
 charcoal. Zinc silicate, for instance, yields no metallic 
 zinc with charcoal, unless it be mixed with a flux capable 
 of combining with the silica. 
 
 4thly. To prevent the formation of alloys of some 
 metals with others, as, for instance, in the case of a mix- 
 ture of the manganese and iron oxides ; when a suitable 
 flux is employed, the iron is obtained in a state of purity, 
 whereas if no flux had been added an alloy would have 
 been obtained. Gold and silver can be separated from 
 many other metals by means of a flux. 
 
 5thly. To scorify some of the metals contained in the 
 substance to be assayed, and obtain the others alloyed 
 with a metal contained in the flux, as gold or silver with 
 lead. 
 
 6thly. A flux may be employed to obtain a single 
 button of metal, which otherwise would be obtained in 
 globules. 
 
 Fluxes are divided into non-metallic and metallic ; the 
 non-metallic fluxes are 
 
 1. Silica. 
 
 2. Lime. 
 
 3. Magnesia. 
 
 4. Alumina. 
 
 5. Calcium and aluminium silicates. 
 
 6. Glass. 
 
 7. Borax (sodium biborate). 
 
 8. Fluor-spar (calcium fluoride). 
 
 9. Potassium carbonate. 
 
192 SILICA, LIME. 
 
 10. Sodium carbonate. 
 
 11. Nitre (potassium nitrate). 
 
 12. Common salt (sodium chloride). 
 
 13. Black flux and its equivalents. 
 
 14. Argol (potassium bitartrate). 
 
 15. Salt of sorrel (potassium binoxalate). 
 
 16. Soap. 
 
 The metallic fluxes are 
 
 17. Litharge (lead oxide) and ceruse (lead carbonate). 
 
 18. Glass of lead (lead silicate). 
 
 19. Lead borate. 
 
 20. Lead sulphate. 
 
 21. Copper oxide. 
 
 22. Iron oxides. 
 
 1. SILICA is employed frequently to cause the fusion of 
 some gangues in assays made at an elevated temperature. 
 Silica combines with all the bases, and forms with them 
 silicates, which are more or less fusible. 
 
 Quartz is the beet form of silica to use. For that pur- 
 pose it must be strongly heated, and then quenched in cold 
 water. It can then be easily pulverised. In case the 
 quartz takes a yellow or reddish colour on ignition, it 
 must be digested with hydrochloric acid. 
 
 2, 3, 4, 5. LIME, MAGNESIA, ALUMINA, AND THEIR SILICATES. 
 No simple silicate is readily fusible, so that lime, magnesia, 
 or alumina are employed, according to circumstances, to 
 reduce a simple silicate to such a condition that it will 
 readily fuse in an assay furnace. Sometimes it may be 
 requisite to use all the above-mentioned earths. 
 
 Pure lime, when exposed to atmospheric air, attracts 
 carbonic acid and water so quickly that, in practice, pure 
 calcium carbonate is used in the form of chalk, calcareous 
 spar, or marble, if they are pure. Calcium carbonate fre- 
 quently contains foreign substances, viz. iron, manganese, 
 alumina, silica, and also magnesium carbonate. A certain 
 quantity of magnesium carbonate is, in many cases, advan- 
 tageous, and alumina and silica are not disadvantageous. 
 
GLASS, BORAX. 193 
 
 Alumina is never used. in the pure state. Washed 
 china-clay which, on burning, becomes white, is used 
 instead. Clay generally contains from 20 to about 40 per 
 cent, alumina, and, if it is used for the formation of sili- 
 cates, a quantitative analysis of its components should 
 first be performed. 
 
 6. GLASS is a very useful flux in certain assays, and, 
 being a saturated silicate, it will serve by itself either as a 
 slag or merely as a covering. The kind employed must 
 contain no easily reducible metallic oxides, and it must 
 especially be free from arsenious acid and lead oxide. 
 
 The subjoined analyses of glass from Bodemann KerVs 
 ProUerkunst will be found useful. (See p. 194.) 
 
 7. BORAX. That kind with 10 atoms .or 47*1 per cent, 
 water effloresces when exposed to atmospheric air ; and the 
 other kind, with 5 atoms or 30 per cent, water, does not 
 effloresce, and crystallises in octahedrons. This difference 
 is immaterial for assaying purposes, but it is of importance 
 in purchasing borax. 
 
 When borax is heated, it loses its water of crystalli- 
 sation and undergoes an enormous increase of volume ; at 
 a higher temperature it fuses and forms a transparent 
 glass, which becomes dull on the surface by exposure to 
 air. Only the fused vitrified borax ought to be used in 
 assays. It must be reduced to powder, and kept in well- 
 closed vessels. 
 
 Borax may be regarded as containing free boracic 
 acid ; it is an excellent and nearly universal flux : it has 
 the property of forming, like boracic acid, fusible com- 
 pounds with silica and nearly all the bases, and is prefer- 
 able to boracic acid because it is much less volatile. It may 
 be used at a high or a low temperature. It is employed 
 in the assay of gold and silver, because it fuses and com- 
 bines with most metallic oxides, or in obtaining a regulus 
 that is to say, to separate the metals, their arsenides and 
 sulphides, from any stony matter with which they may be 
 mixed ; because this salt is neither oxidising nor desul- 
 phurising. It is also employed in the assay of iron and 
 tin ores, as in the presence of charcoal it retains but traces 
 
 o 
 
194 
 
 ANALYSES OP GLASS. 
 
 CQ 
 ft 
 
 H 
 
 FH 
 ^5 
 P^3 
 
 . 
 W 
 
 e 
 
 
 p 
 
 fe 
 o 
 
 QQ 
 
 w 
 
 S 
 
 <j 
 & 
 -< 
 
 I 
 
 ? 9 > *? 9 9 T 1 
 
 XO "tf O rH rH C- O 
 rH 
 
 . CO 
 
 :6 
 
 . 
 : | 
 
 
 
 co 
 
 6 
 jt 
 
 $ 
 
 O 
 
 o 6 
 
 . . C5 C5 . .0 
 
 : : TH o : ' o 
 
 cT 
 fl 
 
 rH 
 
 O 00 CM CO 
 
 OCMrH 
 
 OTH 
 
 03 
 
 
 
 i 
 
 6 
 
 CM 
 
 10 Tin 00 CO . t- CO CM 
 6 CM rH CM C CM CO rH 
 
 CO rH T^ rH O rH <3 rH 
 
 o 
 
 .CM" .cp 
 
 . rH - O 
 
 
 . CM 
 
 
 
 
 
 
 
 
 
 :6 
 
 
 
 
 
 
 
 
 % 
 
 . . CM . 
 
 : CM : 
 
 O O 
 OOCM 
 
 , 
 
 . 
 
 . 
 
 . 
 
 . O CM 
 
 : CM 6 
 
 rH 
 
 
 
 
 
 
 I 1 
 
 CM CO CO ^ 
 (jq (jq rH rH 
 
 O C5 t- O cp Oi t>- . 
 rH rH CM rH ^ 
 
 CO 
 
 rH 
 
 CO 
 
 t- 
 
 rH CO TH CO rH 00 O 
 
 o t* 
 
 rH 
 
 
 
 t" 
 
 00 
 
 rH rH rH 
 
 1 
 
 : cb : 
 
 rH 
 rH 
 
 00 CM 
 
 CO O t- 
 
 rH 
 
 : CM 
 
 CO t^ 
 
 . rH O O . 
 
 : cb ob cb " 
 
 CM 
 
 9 
 
 CO 
 
 s 
 
 00 CM O O CO <* rH 
 
 t- 
 
 (M O t^ CO rH CM X 
 
 
 
 8 
 
 00 C 
 rt< . O O t- O O t- 
 h> '. O OD rH CM >b CM 
 rH rH T 1 
 
 cO 
 
 o t- o 
 
 rH O Tt< 
 CM rH rH 
 
 C5 
 
 2o? 
 rH rH 
 
 $ 
 
 S 
 
 OOrH 
 
 CM 
 
 I-H 
 
 . . . O CO t- Ci 
 
 : : : CO rH rH CO 
 
 o" 
 S 
 
 So o cp 
 
 W3 O CO.C- 
 
 99 
 
 t- 
 
 ft 
 
 
 
 CO 
 rHO 
 
 
 
 CM 
 
 
 
 gcOO^t-OCO 
 
 t>- 1> 
 
 t- t- 
 
 CO t> 
 
 L- 
 L 
 
 
 C~ 
 
 CO 
 
 
 
 . 
 
 Si^^^^g 
 
 
 
 i 
 
 L 
 
 y wine (Warrington) -j 
 
 ^-y-^ 
 
 I 
 
 'Q 
 
 by acids (Struve) 
 (Peligot) 
 inalysis (Rowneyand f 
 
 0) * 
 
 If 
 ii 
 
 is 
 
 r ^. 
 
 11 
 11 
 
 s^ 
 
 
 
 ories in Paris (Salvetat) 
 
 . . 
 
 ) 
 ble 
 
 wine-glass 
 
 ss (Peli 
 decomp 
 
 white hollow 
 
 plate (mirror) gl 
 glass goblet, not 
 
 
 bes for organic 
 
 
 on 
 
 mbus 
 Otto 
 
 Gla goblet from Petersburg, of which 8 
 n extracted by muriatic acid at the 
 French hollow wine-glass ... 
 Frenchglasa for the apparatus of the labor 
 
 gla 
 las 
 
 las 
 
 bee 
 
 l 
 French window 
 
 French mirror-g 
 
 g 
 
 irr 
 ir 
 
 English m 
 
 CMrj5 
 rH rH 
 
 CO 
 
 COt-OO 
 rH rH rH . CM ' 
 
FLUXES FLUOR-SPAR, ALKALINE CARBONATES. 19o 
 
 of their oxides, and, indeed, much less than generally 
 remains with the silicates. 
 
 8. FLUOR-SPAR or CALCIUM FLUORIDE is rarely employed 
 in assays, but in certain cases is an excellent flux, as will 
 be hereafter shown. 
 
 9, 10. POTASSIUM CARBONATE and SODIUM CARBONATE. It 
 has been already shown that they possess oxidising and 
 desulphurising power ; they will now be considered as 
 fluxes. 
 
 They are decomposed in the dry way by silica and the 
 silicates, with the separation of carbonic acid. The presence 
 of charcoal much facilitates this decomposition. 
 
 They form fusible compounds with many metallic 
 oxides. In these combinations the oxide replaces a cer- 
 tain quantity of carbonic acid ; but they are not stable 
 they are decomposed by carbon, which reduces the oxides, 
 or by water, which dissolves the alkali. 
 
 On account of their great fusibility, the alkaline car- 
 bonates can retain in suspension, without losing their 
 fluidity, a large proportion of pulverised infusible sub- 
 stances, as an earth, charcoal, &c. 
 
 The alkaline carbonates ought to be deprived of their 
 water of crystallisation for assaying purposes ; in fact, it 
 would be better to fuse them before use. They must in 
 all cases be kept in well-stopped vessels. 
 
 They may be used indifferently, but sodium carbonate 
 is to be preferred, as it does not deliquesce, and is gener- 
 ally much cheaper. 
 
 The alkaline carbonates of commerce always contain 
 sulphates and chlorides. In some cases this causes no in- 
 convenience, but there are many circumstances in which 
 the presence of sulphuric acid would be injurious. 
 
 Potassium carbonate can readily be procured free from 
 sulphate and chloride by means of pure nitre and charcoal, 
 as follows : Pulverise, roughly, 6 parts of pure nitre, and 
 mix with 1 part of charcoal ; then project the mixture, 
 spoonful by spoonful, into a red-hot iron crucible. The 
 projection of each spoonful is accompanied by a vivid 
 deflagration, and potassium carbonate is found in a fused 
 
 o 2 
 
190 FLUXES XITRE, COMMOX SALT. 
 
 state at the bottom of the crucible. It must be pulverised, 
 separated from excess of charcoal, and kept in a dry state 
 for use. 
 
 Sodium carbonate may be obtained in much the same 
 way, substituting sodium nitrate for potassium nitrate. 
 Either carbonate may also be obtained in a sufficient state 
 of purity by repeatedly crystallising the commercial car- 
 bonates. 
 
 11. POTASSIUM NITRATE. Its properties have already 
 been pointed out. The presence of silica or of silicates 
 much assists its decomposition. 
 
 12. COMMON SALT (SODIUM CHLORIDE) is recommended 
 either mixed with flux, or placed above it, for the purpose 
 of preserving the substance beneath from the action of 
 the atmosphere, or to moderate the action of such bodies 
 as cause much ebullition. It is very useful in lead assays, 
 and is much used in the assay of silver by the wet way. 
 It must be previously pounded, and heated to dull redness 
 in a crucible, to prevent its decrepitation. 
 
 Common salt, though containing calcium and magne- 
 sium sulphates and chlorides, is in most cases sufficiently 
 pure for assaying purposes. If intended for copper assays 
 it must be previously purified from sulphates. 
 
 Plattner * has examined the influence of common salt 
 upon different oxides and sulphates. It does not act upon 
 uncombined lead and zinc oxides. Lead sulphate, when 
 melted with it at a dull red heat, becomes liquid, and 
 evolves vapour of lead chloride. By raising the tempera- 
 ture, and by giving more draught of air, the evolution of 
 such vapour is increased. Common salt acts upon zinc 
 sulphate in the same way. Antimony oxide and antimo- 
 nious acid heated with it at a dull red heat evolve vapour 
 of antimony chloride, though not in a great quantity. 
 Copper sulphate melted with salt at a red heat becomes 
 converted into copper chloride and sodium sulphate. 
 Copper chloride becomes vaporised if air is admitted, and 
 it becomes converted into copper subchloride by raising 
 the temperature a little, chlorine being then evolved. 
 
 * <B. u. h. Ztg.' 1854, p. 126. 
 
WHITE, BLACK, AND RAW FLUX. 197 
 
 13. WHITE FLUX, BLACK FLUX, and RAW FLUX White flux 
 
 Is produced by deflagrating together equal parts of salt- 
 petre and argol (crude potassium bitartrate) ; black flux, 
 by deflagrating one part of saltpetre with two or three or 
 more parts of argol. Generally one part of saltpetre and 
 two and a half parts of argol are taken. The finely pul- 
 verised and intimate mixture for either flux, before it is 
 deflagrated, is called raw flux. 
 
 After the saltpetre and argol have been finely pulver- 
 ised and sifted separately, they are intimately rubbed to- 
 gether, and then deflagrated by throwing the mixture little 
 by little into a low-red-hot crucible, which after each 
 addition is lightly covered over. The deflagration may 
 also be conducted, though less advantageously, by filling 
 the crucible about two-thirds full of the raw flux and then 
 touching it with a red-hot coal or iron. It can only be 
 performed in the open air or under a flue with a strong 
 draught, as the tartaric acid evolves various empyreumatic 
 volatile matters in considerable quantity during its decom- 
 position. 
 
 With white flux the saltpetre suffices to burn all the 
 charcoal produced by the carbonisation of the tartaric 
 acid, and the result is therefore almost pure potassium 
 carbonate, if pure saltpetre and pure argol have been used. 
 If the latter was impure, the resulting neutral potassium 
 carbonate may contain much, perhaps 10 per cent., of cal- 
 cium carbonate. White flux works like ordinary potas- 
 sium carbonate, which is therefore almost always preferred 
 to the far more expensive flux. 
 
 With the black flux the quantity of saltpetre is not 
 sufficient to burn all the carbon from the argol, and there 
 remain therefore in the black flux, according as two, two 
 and a half, or three parts of argol were taken, about 5, 
 8, or 12 per cent, of free carbon, which is mixed in the 
 most intimate manner with the resulting neutral potassium 
 carbonate more intimately, indeed, than would be possible 
 by any mechanical means. This charcoal does not hinder 
 the fusing of the assay when the flux is used, and effects or < 
 promotes the reduction of the metallic oxides. 
 
1 ( J8 WHITE, BLACK, AND RAW FLUX. 
 
 Fusion and reduction, sometimes also desulphurisation.. 
 are the purposes for winch black flux is used, and, accord- 
 ing to the special character of the assay, a greater or a less 
 proportion of charcoal to the carbonate of potash may be 
 desirable, and this is to determine whether two, two and 
 a half, three, or more parts of argol are to be used to one 
 of saltpetre. As a general rule it may be stated, the more 
 difficultly fusible is the assay, the more potash must be 
 present ; and the more metallic oxide is to be reduced, 
 the more charcoal ; and the more also of the latter, the 
 more oxygen the oxide contains. 
 
 In many cases, instead of black flux, a mixture of po- 
 tassium carbonate and powdered charcoal, in a suitable 
 ratio to each other, suffices, especially if the mixture, before 
 use, is passed through a sieve, or otherwise very intimately 
 mingled. Instead of the powdered charcoal, also, a corre- 
 sponding (about twice or four times as large) quantity of 
 flour, sugar, or starch may be mixed with the potassium 
 carbonate. Lamp-black is, however, the best form of car- 
 bon. The three following fluxes are very useful : 
 
 Sodium carbonate 94 88 816 
 
 Charcoal . . . . . .6 12 184 
 
 The second is very nearly equivalent to sodium and 
 carbonic acid, and the third to sodium and carbonic oxide ; 
 but it must be observed that, whatever precautions be 
 taken, these mixtures never become so liquid as black 
 flux, because the charcoal tends very much to separate and 
 rise to the surface. 
 
 A mixture of 100 parts .of pure potassium carbonate 
 and 10 to 15 parts of wheat or rye flour is to be preferred 
 to black flux in case the argol contains gypsum, or the 
 saltpetre, sulphates, which in many cases might work in- 
 juriously upon the assay. If this is the case, then, in the 
 presence of a reducing flux, sodium sulphide is apt to form, 
 which, for example in the copper assay, occasions the 
 slagging of copper. - 
 
 Cream of tartar, carbonised by a semi-combustion until 
 it is reduced to half its weight, is a very good substitute 
 for black flux; it contains about 10 per cent, of charcoal. 
 
FLUXES CREAM OF TARTAR. 
 
 As a perfectly general rule for the use of black flux, 
 and of mixtures similar to it, it is to be observed that the 
 crucible should never be more than two-thirds filled, as 
 the assay always intumesces, i.e. evolves gaseous matters, 
 when free carbon is present. 
 
 14. ARGOL, CREAM OF TARTAR, or POTASSIUM BITARTRATE. 
 When potassium bitartrate is heated in a covered crucible, 
 a rapid decomposition takes place, accompanied by a dis- 
 engagement of inflammable gases ; the substance agglo- 
 merates, but without fusing or boiling up. The residue 
 is black and friable, and contains 15 per cent, of carbon 
 when produced from rough tartar or argol, and 7 per 
 cent, from cream of tartar, 
 
 These reagents produce the same effects as black flux, 
 and possess more reducing power, because they contain 
 more combustible matter : but this is an inconvenience, 
 for the excess prevents their entering into full fusion when 
 the substance to be assayed requires but a small propor- 
 tion of a reducing agent. They can be used with success 
 in assays requiring much carbonaceous matter. 
 
 15. SALT OF SORREL, or POTASSIUM BINOXALATE, when 
 heated, is decomposed. It decrepitates feebly, and during 
 its decomposition is covered with a blue flame ; it at first 
 softens, and when fully fused is wholly converted into car- 
 bonate. When the oxalate is very pure, the resulting car- 
 bonate is perfectly white, and free from charcoal ; but 
 very often it is spotted with blackish marks. It has no 
 very great reducing power. 
 
 16. WHITE or MOTTLED SOAP is a compound of soda 
 with a fatty acid. When heated in closed vessels it fuses, 
 boiling up considerably, and during its decomposition 
 gives off smoke and combustible gases, and leaves a residue 
 composed of sodium carbonate with about 5 per cent, of 
 charcoal. Of all reducing agents, soap absorbs the great- 
 est quantity of oxygen ; and as the residue of its decom- 
 position by heat affords but little charcoal, it has the 
 property of forming very fluid slags. Nevertheless it is 
 rarely employed, because certain inconveniences outweigh 
 its advantages. These inconveniences are, its bubbling up, 
 
 
200 REDUCING TOWER OF FLUXES. 
 
 and its extreme lightness. It also requires to be rasped, in 
 order to mix it perfectly with the substances it is to decom- 
 pose, and it then occupies a very large volume, and requires 
 correspondingly large crucibles. By mixing rasped soap 
 with potassium binoxalate or sodium carbonate excellent 
 reducing fluxes may be made. 
 
 Reducing Power of the Various Fluxes. By fusing 
 equal weights of each of the above-mentioned reducing 
 fluxes with an excess of litharge, the following quantities 
 of lead were yielded : 
 
 Common black flux, made with two parts of tartar . . 1'40 
 
 Ditto, with 2 of tartar 1'90 
 
 Ditto, with 3 of tartar . . . >>. 3 ' 80 
 
 Sodium carbonate 
 
 Charcoal . 
 
 Sodium carbonate u L 
 
 Charcoal . *" * 
 
 Sodium carbonate c/v i 
 
 Sugar . 10 / 
 
 Sodium carbonate 80 1 
 
 Sugar . . ^ 2 ' 8( 
 
 Sodium carbonate 
 
 Starch 
 
 Sodium carbonate 80 ) aA 
 
 Starch > ' ' ' ' ' ' 2<d 
 
 Crude tartar, argol 
 Cream of tartar . 
 
 5-60 
 4-50 
 
 Salt of sorrel . 851 or 
 
 Soap ... 15 / ' ' ' ' ' 
 
 Sodium carbonate 85 1 
 
 Soap ... 15/ 
 
 Cream of tartar, carbonised ...... 3*10 
 
 Ditto, ditto, calcined 2-20 
 
 Potassium binoxalate '90 
 
 White soda soap . . 16-00 
 
 All fluxes containing alkaline and carbonaceous sub- 
 stances are reducing and desulphurising, besides acting as 
 fluxes properly so called. They also produce another 
 effect which it is useful to know, viz. they 'have the pro- 
 perty of introducing a certain quantity of potassium or 
 sodium into the reduced metal. This was first pointed out 
 by M. Vauquelin.* He found that when antimony, bis- 
 muth, or lead oxide was fused with an excess of tartar, the 
 metals obtained possessed some peculiar characters, which 
 they owed to the presence of potassium. 
 
 * ' Annales des Mines.' 
 
METALLIC FLUXES. 201 
 
 METALLIC FLUXES. 
 
 IT. LITHARGE AND CERUSE. These bodies always act as 
 fluxes, but at the same time often produce an alloy with 
 the metal contained in the ore to be assayed. Ceruse 
 produces the same fluxing effect as litharge. The former 
 is the better flux, and is very useful in a great number of 
 assays. 
 
 18. GLASS OF LEAD (LEAD SILICATE). Lead silicates are 
 preferable to litharge in the treatment of substances con- 
 taining no silica, or which contain earths or oxides not 
 capable of forming a compound with lead oxide excepting 
 by the aid of silica. It may be made by fusing 1 part of 
 sand with four parts of litharge : if required more fusible, 
 a larger proportion of litharge must be added. 
 
 19. LEAD BORATE. The lead borates are better fluxes 
 than the silicates when the substance to be assayed con- 
 tains free earths ; but in order to prevent them swelling 
 up much when fused, they must contain an excess of lead 
 oxide. The lead borate containing 90*56 of lead oxide 
 and 9 -44 of boracic acid is very good. Instead of lead 
 borate, a mixture of fused borax and litharge may be 
 employed ; it is equally serviceable. 
 
 20. LEAD SULPHATE is decomposed by all siliceous 
 matters, and by lime, so that when these substances are 
 present litharge is produced, which fluxes them. 
 
 21. COPPER OXIDE is rarely used as a flux for oxidised 
 matters, but is sometimes employed in the assays of gold 
 and silver, to form an alloy with those metals. In this 
 case a reducing flux must be mixed with the oxide. 
 Metallic copper may be used, but is not so useful, as it 
 cannot be so intimately mixed with the assay. 
 
 22. THE IRON OXIDES are good fluxes for silica and the 
 silicates. They are, however, rarely employed for that 
 purpose ; they are more often used to introduce metallic 
 iron into an alloy to collect an infusible, or nearly infusible, 
 metal, by alloying it with iron ; such as manganese, 
 tungsten, or molybdenum. 
 
202 
 
 CHAPTEE VII. 
 
 THE BLOWPIPE AND ITS USE. 
 
 THE blowpipe was formerly only used by jewellers and 
 workers of metal for producing sufficient heat for solder- 
 ing certain small portions of their work ; and it was not 
 till about the year 1733 that Anton Swab applied it to 
 the analysis of mineral substances. Cronstedt used the 
 blowpipe to ascertain the difference between various mine- 
 ral substances as to fusibility, &c. In 1765 Yon Enge- 
 strom published Cronstedt's ' System of Mineralogy,' and 
 added to it a ' Treatise on the Blowpipe,' in which he 
 pointed out the process of Cronstedt. 
 
 This work attracted the attention of philosophers to 
 this valuable instrument, and its use became more general, 
 and was further developed by Bergman and Gahn. 
 
 Berzelius, after Gahn, was particularly famed for his 
 skill with the blowpipe, and for his improvements in the 
 form of apparatus ; and from his excellent work on this 
 subject some of the following descriptive part of Blow- 
 pipes, Lamps, Tongs, &c., is derived. 
 
 The common blowpipe of gasfitters, jewellers, &c., is 
 a tube of brass, tapering towards one end, and curved at 
 that extremity, which has an opening as fine as that made 
 by the finest needle; it is this opening which is held 
 against the flame of the lamp, and air is blown to it to 
 increase the amount of heat. In all ordinary operations 
 the blast is required to be kept up not more than a 
 minute, so that the quantity of moisture exhaled from the 
 lungs produces no inconvenience by stopping up the tube. 
 But in certain chemical operations this is exceedingly 
 troublesome, as a continuous blast is required, and a large 
 
THE BLOWPIPE AND ITS USE. 203 
 
 quantity of water collects in consequence, generally suffi- 
 cient to mar the success of an experiment. In order to 
 obviate this, Cronstedt placed in the centre of his blow- 
 pipe a bulb, in which the greater part of the water col- 
 lected. This form was, however, inconvenient, because 
 if the jet of the blowpipe were at all inclined, even for an 
 instant, the water ran from the bulb and filled it. In a 
 series of articles communicated to the ; Chemical News,' 
 Mr. David Forbes, F.K.S., has given directions which are 
 invaluable to all who practise with this instrument. From 
 these we quote the following. 
 
 ' BLOWPIPE. The form adopted long ago by Gahn is con- 
 sidered as the most convenient. FIG 67 
 Fig. 67 shows an improvement 
 made by the author upon this 
 form. 
 
 ' In this figure it will be seen 
 that the arm of the jet is 
 double, turning upon a central 
 hollow axis, which allows the 
 blast to be directed at will 
 through either half of the arm, 
 merely by rotating the arm it- 
 self half round ; by having con- 
 sequently the two holes with 
 
 respectively a large and small 
 
 orifice, a corresponding blast 
 
 may be obtained at pleasure, 
 
 without suspending the opera- 
 tion. 
 
 ' As a more steady and 
 
 long-continued blast is required 
 
 in quantitative operations than 
 
 could be kept up by using a 
 
 blowpipe provided with an 
 
 ordinary mouthpiece held be- 
 tween the lips without seriously 
 
 distressing the muscles of the 
 
 cheeks, it is quite essential that the trumpet mouthpiece 
 
204 MR. FORBES'S BLOWPIPE: 
 
 shown in fig. 67 be adopted ; for the same reasons also 
 the mode of holding the blowpipe represented in fig. 68 is 
 FlG 68 recommended, as securing the greatest 
 
 steadiness from motion, and as greatly 
 assisting the muscles of the cheeks 
 by the external support afforded 
 them by the position of the thumb 
 pressing against the trumpet mouth- 
 piece. 
 
 ' The nipples are turned, and bored 
 
 of three different sizes, and are made both of platinum 
 and of brass. The first, of platinum, contains the smallest 
 .aperture, and is employed for qualitative analysis ; the 
 second, of brass, is used for such qualitative experiments 
 as require a strong oxidising flame, and for heating silver, 
 gold, and copper, in quantitative assay ; also for roasting 
 copper, lead, and tin ores, the metallic contents of which 
 are to be accurately determined ; and the third, which is 
 .also manufactured of brass, has the largest bore, and is 
 used for the quantitative estimation of lead and tin. 
 
 ' Platinum nipples are, however, always preferable to 
 those of brass, because by exposure to a moderate red 
 heat on charcoal before the blowpipe they are more easily 
 cleaned from the sooty particles which obstruct the aper- 
 ture. This method of cleansing cannot be applied to brass 
 nipples, owing to their rapid oxidisation ; to clean these 
 the operator must adapt in the opening a sharp-pointed 
 fragment of horn, or a small needle, ground along one- 
 half of its length ; by this means the aperture throng] i 
 which the air passes may be readily cleansed.' 
 
 The power and perfection of the blowpipe flame greatly 
 depends upon the internal construction of these jets or 
 nipples. The current of air for at least three-quarters of 
 an inch of the orifice should meet with no obstruction or 
 roughness such as a screw thread or angle so frequently 
 met with in blowpipes having removable nozzles. The 
 most perfect form for the blowpipe jet is that obtained by 
 slightly thickening and drawing down a piece of glass 
 tubing of about 5 millims. internal diameter to the required 
 
BLOWPIPE JETS. 205 
 
 size for the jet, then cutting the contracted part cleanly 
 across at a point about half an inch from one of the 
 shoulders. We then have a jet of the shape shown in 
 
 FIG 69. 
 
 fig. 69, with which the most beautiful flame can be pro- 
 duced, and almost any desired pressure of air used without 
 hissing. 
 
 It is needless, however, to remark that the liability of 
 glass jets to crack and fuse renders their use impracticable 
 except for an oxidising flame, where the jet need not be 
 inserted in the flame of the lamp. Metal, therefore, must 
 be resorted to, and, as the superiority of the glass jet? 
 solely depends on their internal shape and smoothness, the 
 nearer the metal jets can be made to approach them the 
 greater will be the satisfaction in using the blowpipe. 
 
 The jets should be at least half an inch long and 
 
 FIG. 70. 
 
 coned into the blowpipes, as shown in fig. 70, not screw r ed 
 as is generally the case. 
 
 To obtain, therefore, the necessary internal shape for 
 satisfactory blowpipe jets, too much care cannot be taken 
 over the work of drilling, which operation can only be 
 successfully performed by means of a specially constructed 
 drill. 
 
 Any kind of flame may be used for the blowpipe, pro- 
 vided it be not too small ; a candle, a lamp, or gas may 
 be employed ; Engestroni and Bergman used common 
 candles in preference. Berzelius employs a lamp, which 
 is certainly much preferable to a candle. I have occasion- 
 ally employed the flame of coal gas, which answers very 
 well, but is not so good as that of a lamp. Berzelius says 
 on this subject, ' Lamps have doubtless many advantages 
 
206 USING THE BLOWPIPE. 
 
 over candles, but are not so convenient in travelling, on 
 account of the escape of oil. The oil employed ought to 
 be the best olive or salad oil.' 
 
 The lamp recommended by Mr. Forbes has the ad- 
 vantage of being portable, and closes in such a manner 
 that no oil can escape. It is made of japanned tin-plate, 
 and is about 4 inches long and 1 inch wide, furnished at 
 one end with a wick-holder, capable of being completely 
 closed by a screw, and at the other with a ring of tin-plate, 
 which passes over the upright end of a support. It may 
 be mentioned that the screw cap is furnished with a 
 leather washer, by the aid of which it can be rendered 
 much tighter, and the escape of oil entirely prevented. 
 
 Mr. Forbes says that olive oil, burnt in the usual Ber- 
 zelius blowpipe lamp, is probably superior to any other. 
 Gas is not to be recommended, as it is difficult to obtain a 
 good reducing flame when using it. For cupellation and 
 such other operations, however, which only require an 
 oxidising flame, it is excellent. 
 
 A spirit lamp may sometimes be used in blowpipe 
 assays, particularly when glass tubes are employed, as in 
 the detection of volatile substances. In these cases it is 
 much more convenient ; as an oil lamp, in the first place, 
 blackens the tube ; and secondly, does not yield sufficient 
 heat, except when the blowpipe blast is employed. 
 
 It is very difficult to describe in writing a method 
 whereby a student may acquire the practice of using the 
 mouth-blowpipe; that given by Faraday * is perhaps the 
 clearest and most concise. He says, ' The practice neces- 
 sary, in the first place, is that of making the mouth replace 
 the lungs for a short time, by using no other air for the 
 blowpipe than that contained in it.' This practice is sim- 
 ple in itself, and easy to acquire, but, as before stated, 
 difficult to describe. Let the student first observe that it 
 is easy after having closed the lips to fill the mouth with 
 air, and to retain it so, at the same time respiration may 
 be carried on ; and if, while the mouth is in this state, a 
 blowpipe be introduced between the lips, it will be found 
 
 * ' Chemical Manipulation.' 
 
USING THE BLOWPIPE. 207 
 
 that the small quantity of air which its jet allows to pass 
 through it will be amply supplied for ten or fifteen 
 seconds by the quantity contained in the mouth ; and this 
 being repeated a few times, a ready facility for using the 
 blowpipe, independent of the lungs, will be acquired. 
 
 This step being taken, the next is to combine this pro- 
 cess with the ordinary one of propelling air directly from 
 the lungs through the mouth, in such a way that when 
 the action of the lungs is suspended during inspiration, 
 the blast may be continued by the action of the mouth 
 itself, from the air contained within it. The time of four- 
 teen or fifteen seconds, during which the mouth can supply 
 air independently of the lungs, is far more than that re- 
 quired for one or even many more inspirations ; and all 
 that is required to acquire the necessary habit is the power 
 of opening and closing the communication between the 
 mouth and the lungs, and between the air and the lungs, 
 at pleasure. 
 
 The capability of closing the passages to the nostrils is 
 very readily proved : every one possesses and uses it when 
 he blows from the mouth, and that of closing or opening 
 the mouth to the lungs may be acquired with equal readi- 
 ness. Applying the blowpipe to the lips as before, use the 
 air in the mouth to produce a current, and, when it is 
 about half expended, open the lungs to the mouth, so as 
 to replace the air which has passed through the blowpipe ; 
 again cut off the supply, as at first, but continue to send 
 a current through the instrument, and, when the second 
 mouthful of air is gone, renew it as before from the lungs. 
 
 To some this may be difficult ; b ut if the preceding 
 instructions be followed and persevered in for a short time, 
 the learner will soon find that he can keep up a continuous 
 blast from ten minutes to a quarter of an hour, without 
 any other inconvenience than the mere lassitude of the 
 lips caused by compressing the mouthpiece of the instru- 
 ment, and this may be avoided by using the trumpet 
 mouth-piece as recommended by Mr. Forbes. 
 
 After having conquered the difficulty of keeping up a 
 continuous blast, the student must learn how to attain the 
 
208 REDUCTION AND OXIDATION. 
 
 maximum of heat with the least exertion to himself. The 
 chief points to be observed are, neither to blow too 
 fiercely nor too gently : in the first case, the force of the 
 blast would carry away heat by the quantity of cold air 
 thrown into the flame ; and, in the second, a sufficient 
 amount of heat would not be obtained, because a less 
 amount of air would pass into the flame than that required 
 for perfect combustion. 
 
 The highest degree of temperature is required in test- 
 ing the fusibility of many bodies, as also in the reduction 
 of certain oxides, as those of iron, tin, &c. We have yet 
 another class of phenomena to describe, which do not 
 essentially depend on a high temperature ; these are the 
 processes of reduction and oxidation. In order to explain 
 and point out the best methods of effecting these two 
 objects, it will be necessary to enter somewhat into the 
 nature of flame ; this will be done as briefly as is consis- 
 tent with perspicuity. The species of flame examined 
 will be that of a candle, as it is a similar one to that 
 with which the blowpipe operator will have to experi- 
 ment. 
 
 On careful examination it will be found that the flame 
 of a candle or lamp may be divided into four distinct por- 
 tions : first, a deep blue ring at the base ; this consists of 
 the vapour of the combustible, which can hardly burn 
 because it has not acquired a sufficient temperature ; 
 secondly, a dark cone in the centre ; this is also the vapour, 
 but heated intensely, not, however, in a state of combus- 
 tion, on account of the absence of air ; thirdly, of a very 
 brilliant envelope, which surrounds the dark parts just 
 mentioned ; this is the partially consumed vapour at a very 
 high temperature ; the luminous property it possesses is 
 due to the precipitation and subsequent ignition of par- 
 ticles of solid carbon ; and fourthly, of an almost invisible 
 envelope which surrounds the luminous portion ; this is 
 the substance of the combustible in full ignition, it here 
 mingles with the atmospheric oxygen, and is consumed. 
 The highest degree of temperature in the whole flame is to 
 be found at the point of contact between the luminous and 
 
KEDUCTIOX AXD OXIDATION. 209 
 
 this part. It must be particularly borne in mind that the 
 inner portions of the flame have an excess of carbonaceous 
 matters, and the outer an excess of oxygenated matters. 
 
 Having premised this much, we will examine the nature 
 of the flame of a candle when acted on by the blowpipe 
 blast, and ascertain how far it is altered, and what are the 
 properties of its separate parts in relation to their oxidis- 
 ing and reducing powers. Supposing the lighted lamp or 
 candle be ready and neatly snuffed, place the nozzle of the 
 blowpipe just in the edge of the flame, and about the six- 
 teenth of an inch above the level of the wick : when 
 things are in this state, blow gently and evenly through 
 the blowpipe, and a conical jet or dart of flame will be 
 produced, which, when formed in a steady atmosphere, 
 free from accidental draughts and currents, will be found 
 to consist of two essential parts the inner cone, blue, 
 small, and well defined ; the outer, brownish and vague. 
 The greatest intensity of heat is found a little beyond the 
 apex of the blue flame ; it is there, also, reduction takes 
 place. The outer flame is formed by the complete com- 
 bustion of the combustible matter of the inner ; and at that 
 place, and just beyond it, oxidation takes place. 
 
 Oxidation^ as before stated, takes place at the extremity 
 of the outer flame, hence it is termed the oxidising flame ; 
 in it all the combustible portions are supersaturated with 
 oxygen. In general the further the substance to be oxidised 
 can be placed from the extremity of the flame, the better 
 the operation proceeds, provided always that the necessary 
 temperature be maintained. Dull redness is the best suited 
 for oxidation. 
 
 Reduction. In this operation the jet of the blowpipe 
 must be introduced into the body of the flame, so as only 
 to produce a small dart ; and a jet having a smaller hole 
 than that used for oxidation ought to be employed. By 
 operating thus a more brilliant flame than the last is pro- 
 duced ; it is the result of a less perfect combustion, and 
 therefore contains a large amount of carbonaceous matter, 
 fitting it more especially for the purpose of separating 
 oxygen from all metallic bodies. 
 
 p 
 
210 AUXILIARY BLOWPIPE APPARATUS. 
 
 Berzelius says, ' The most important point in blowpipe 
 assays is the power of producing oxidation and reduction 
 at will.' Oxidation is so easy, that to do it requires only 
 to read a description of it ; but reduction requires some 
 practice, and a certain knowledge of producing various 
 kinds of blasts. One of the best methods of exercise in 
 this operation is to take a small grain of tin, and place it 
 on charcoal ; then direct the blowpipe dart upon it it 
 will soon fuse ; and if the operator has not produced a 
 good reducing flame, it will become covered w r ith a crust 
 of oxide ; so that it becomes a witness against him each 
 time this happens. The nature of the flame must be altered 
 until, by observation, the proper kind is produced at will. 
 The longer the button of tin is kept bright, the better and 
 more expert the operator. 
 
 AUXILIARY BLOWPIPE APPARATUS, ETC. 
 
 Supports. The support is the substance destined to 
 hold the material to be assayed whilst under the influence 
 of heat. From this it will be seen that an exceedingly 
 refractory body must necessarily be employed, so as not 
 to give way under the excessive heat ; and also (with the 
 exception of charcoal) it ought to have no chemical action 
 on the substances placed in contact with it. Supports 
 may be divided into combustible and incombustible : 
 the former is charcoal ; and for the latter, metal, glass, 
 and earthenware, and in some cases certain minerals, have 
 been employed. 
 
 Charcoal. Mr. Forbes gives the following excellent 
 description of the preparation of charcoal for blowpipe 
 purposes : c lt is extremely difficult to obtain, in England, 
 charcoal fit for blowpipe operations without special pre- 
 paration. The charcoal sold is generally of hard wood, 
 badly burnt, full of cracks, and decrepitating upon appli- 
 cation of heat. Good charcoal should be soft, yet com- 
 pact, and without cracks, and is best made from fir or 
 pine. Where good charcoal cannot be obtained, it can be 
 made artificially by moulding charcoal powder agglutinated 
 
CHAKCOAL APPARATUS. 211 
 
 by some starch paste, and, after desiccation, burning the 
 pieces in a crucible filled with sand. 
 
 ' For the preparation of the charcoal used as a support 
 for the assays, the instruments represented in fig. 71 are 
 required, all of which are fitted in the universal handle a, 
 which is shown in this figure p IG> 71< 
 
 as holding the largest charcoal 
 borer, a section and plan of 
 which are shown in b. This 
 large borer is employed for 
 forming the deep holes in the 
 charcoal used in the blowpipe 
 furnace, and which serve to 
 contain the clay crucibles or 
 capsules in which the assays 
 are fused. The blast-holes in 
 the charcoal inside the blowpipe 
 furnace are bored out by the 
 gouge-shaped borer d, which 
 also serves for making small 
 holes or grooves in charcoal for 
 general purposes. The smaller 
 borer c is most useful, particularly in boring out the holes 
 for receiving the soda paper cornets containing the assay 
 for reduction. The saw-knife e also fits into the same 
 handle, and is used for trimming and sawing across the 
 charcoal pieces, having coarse saw-teeth in front, whilst 
 the back presents a sharp knife-edge. The figures are 
 all drawn to one-half of the real size.' 
 
 Platinum. This metal is much employed as a support 
 in cases where charcoal would be injurious by its reducing 
 power. It is used in three forms, viz. wire, foil, and as a 
 spoon or small capsule. 
 
 Wire. A moderately strong wire of platinum, about 2 
 inches long, and curved at one end, is used with great 
 advantage in many quantitative examinations. The curve 
 serves as a support in all experiments on tests of oxidation 
 and reduction, where alteration of colour only is to be 
 observed. This support can be relied on, for it is totally 
 
 p 2 
 
212 CHARCOAL APPARATUS. 
 
 free from the false varieties of colour which are too often 
 perceptible when the assay rests on charcoal. In the 
 treatment of metals, or in reduction tests, where an easily 
 melted body is to be operated upon, charcoal must, how- 
 ever, be used. It is necessary to have at hand several 
 platinum wires, so as to proceed to another experiment 
 without being obliged to forcibly remove the adhering 
 borax glass, or to wait for its solution in hydrochloric 
 acid, which is the better mode. If the platinum loop 
 melts with the reagent, it must be cut away, and a new 
 one formed. A wire can be used for a very long time, 
 and when it becomes too short to be held between the 
 fingers the straight end may be fastened into a cork, or a 
 piece of glass tubing. 
 
 The platinum spoon (see fig. 72) and foil are used in 
 much the same way ; but as charcoal and the platinum wire 
 FIG. 72. answer every purpose, it 
 
 will be unnecessary to de- 
 scribe their use further : 
 small iron spoons of the 
 above form are also made, 
 and are very useful in 
 cases where the presence of 
 iron is not objectionable. 
 Other instruments, as forceps, hammer, anvil, agate 
 mortar, scissors, &c., are sufficiently familiar to every- 
 body not to require description. Special apparatus 
 required for any operation will be described in the course 
 of the processes. 
 
 Colonel Eoss recommends as a support in certain cases 
 a thin rectangular slip of aluminium plate, 4 inches by 2, 
 half an inch of the lower end of which is turned up at an 
 angle of 80 as a rest for the assay. A fragment of the 
 substance of the size of a pea is laid upon the edge close 
 to the angle, and heated very slightly about half an inch 
 from the top of a pure blue flame. The sublimate obtained 
 should be examined from time to time, increasing the heat 
 after each examination, till nothing more is obtained. 
 For a full description of this method, by means of which 
 
BLOWPIPE EEAGENTS AND FLUXES. 213 
 
 substances existing in mixture or combination may be 
 usefully separated by taking advantage of their different 
 degrees of volatility, the reader is referred to 'Alphabetical 
 Manual of Blowpipe Analysis ' (Triibner & Co.) 
 
 Asbestos cardboard, according to Mr. W. M. Hartley, 
 may also serve for supports. This substance resembles 
 greyish cardboard, but has a soapy feel, like steatite ; it 
 can be used for making crucible supports, sand-baths, 
 muffles, retort supports, &c. ; it can be cut with cork 
 borers or scissors ; by moistening with water it can be 
 moulded to any shape. After moistening it should be 
 gradually dried and ignited, to get rid of organic matter. 
 It stands the ordinary wear and tear of the laboratory 
 well. It is formed principally of asbestos fibres. It can 
 be obtained from the manufactory, 31 St. Vincent Place, 
 Glasgow, at 4s. a pound. 
 
 REAGENTS AND FLUXES. 
 
 BLUE LITMUS PAPER is used for detecting free acid in 
 solution, its colour being changed to red. 
 
 REDDENED LITMUS PAPER is used for the detection of free 
 alkali, its colour being restored to blue. 
 
 BRAZIL-WOOD PAPER is used for detecting hydrofluoric 
 acid, being tinged straw-yellow when immersed in a very 
 dilute solution of this acid. 
 
 TURMERIC PAPER is used for detecting free alkalies : the 
 change produced is very characteristic, its bright yellow 
 colour becoming dark brown. 
 
 NITRIC ACID is employed in the solution of various 
 metals, alloys, and ores, and for the discrimination of 
 certain precipitates ; also as an oxidising agent. 
 
 ZINC is principally employed for the reduction of 
 antimony and tin. 
 
 COPPER is used for the reduction of mercurial salts and 
 for the detection of arsenious acid. It is also used to 
 detect the presence of nitric acid. 
 
 IRON WIRE is employed to precipitate many metals, and 
 in the separation of sulphur and the fixed acids from any 
 
214 BLOWPIPE REAGENTS AND FLUXES. 
 
 substance with which they may be combined. The metals 
 which can thus be precipitated, or deprived of sulphur, 
 are copper, lead, nickel, and antimony. For instance, if a 
 small piece of iron (pianoforte) wire be placed in a sub- 
 stance in fusion, and acted upon by the blowpipe, it 
 becomes covered with the reduced metal. The latter 
 sometimes appears as small globules. 
 
 Iron has the property of reducing phosphorus from 
 phosphoric acid or the phosphates, giving rise to a phos- 
 phide of iron, which forms on fusion a white, brittle, 
 metallic globule. 
 
 POTASSIUM CYANIDE. This is a most useful flux. MM. 
 Haidlen and Fresenius say, ' We have examined its action 
 on many oxides, sulphides, salts, &c., in reference to its 
 use as a reagent combined with the blowpipe. We prefer, 
 in general, a mixture of equal parts of anhydrous soda 
 and potassium cyanide. This mixture was employed on 
 account of the great facility with which the pure cyanide 
 fuses. In acts, in general, so very similarly to pure soda, 
 that it would be superfluous to describe singly the changes 
 which each individual body appeared to undergo when 
 exposed to its action. We cannot, however, pass over the 
 following especial advantages which it possesses as com- 
 pared with soda. First, reductions are obtained with such 
 great facility that the least practised operator may execute 
 reductions which would otherwise be very difficult : for 
 instance, the reduction of tin from either its oxide or 
 sulphide ; and secondly, that the fused mixture of potas- 
 sium cyanide with soda is so easily absorbed by the char- 
 coal, that the grains of reduced metal can always be most 
 distinctly perceived, and may be most easily separated 
 therefrom for further examination. 7 
 
 SODIUM CARBONATE, called, for the sake of brevity, Soda. 
 The plain carbonate or the bicarbonate may be in- 
 differently employed ; but in either case it is absolutely 
 necessary that it be free from sulphates. 
 
 There are two objects in view in the employment of 
 soda as an auxiliary to the blowpipe : first, to ascertain if 
 the substances combining with this body be fusible or 
 
REAGENTS IN THE DRY WAY. 215 
 
 infusible ; and secondly, to facilitate the reduction of 
 certain metallic oxides. 
 
 In the fusion of substances with soda there are many 
 things to observe. The necessary quantity must be taken 
 on the moistened point of a knife, and kneaded in the palm 
 of the hand, so that it may form a coherent mass. If the 
 body under examination be pulverulent it must be incor- 
 porated with it, but if in lump it must be placed upon it, 
 forcing it slightly into the moistened soda ; then carefully 
 heated on the charcoal with a gentle flame, until thoroughly 
 dry ; and lastly it may be fused. It generally happens 
 that the soda, at the instant of fusion, is absorbed by the 
 charcoal : but this does not hinder its action on the assay ; 
 for if it be fusible with soda, the latter comes to the surface 
 and attacks it, finally forming a liquid globule. If the 
 substance be infusible in soda, but decomposable by it, it 
 alters its appearance without entering into fusion. But, 
 however, before pronouncing any substance to be infusible 
 by soda, the flux ought to be mixed with the pulverised 
 substance. If in these trials too little soda be taken, a 
 portion of the substance remains solid, and the rest forms 
 a covering of transparent glass ; if too much, the bead of 
 glass becomes opaque on cooling. - It sometimes happens 
 that the assay contains a substance which, being insoluble 
 in the glass of soda, prevents it becoming transparent. 
 Then, in order that we may fall into no error respecting 
 the nature of the glass, it becomes necessary in the two 
 last-mentioned cases to add a new quantity of the body 
 under examination, and then ascertain if a limpid globule 
 cannot be obtained. In general it is the best method to 
 add the soda by successive small doses, and note the 
 changes produced by each addition. It sometimes happens, 
 in this kind of assay, that the glass becomes coloured at 
 the moment of cooling, and finally takes a yellow or deep 
 hyacinth-red ; it even becomes occasionally opaque and 
 yellowish-brown. These phenomena indicate the presence 
 of sulphur, either in the assay or the soda employed. If 
 the same colour be constantly produced by the same soda, 
 it is a proof that it contains sulphate of soda ; it must then 
 
2K5 SEDUCTION BEFORE THE BLOWPIPE. 
 
 be discarded ; but if it gives generally a colourless glass, it 
 is the substance under assay that contains sulphur or sul- 
 phuric acid. 
 
 Reduction of Metallic Oxides. This species of assay, 
 by which quantities of reducible metals, so small as to 
 escape ordinary wet analyses, can be detected, is the most 
 important discovery Gahn made in the application of the 
 blowpipe. 
 
 If a small quantity of native or artificial oxide of tin 
 be placed on charcoal, it requires a long blast and a skil- 
 ful operator to produce a grain of metallic tin ; but if a 
 small quantity of soda be added, the reduction takes place- 
 readily, and so completely with pure oxide, that the whole 
 is transformed into a button of tin. From this it is 
 certain that the presence of soda favours the decom- 
 position. 
 
 The action, however, can be explained thus, as Berzelius 
 himself hints : the red-hot charcoal reacts upon the sodium 
 carbonate, producing by its reduction a certain amount of 
 sodium, which by its strong attraction for oxygen seizes on 
 that contained by the metallic oxide which is required to 
 be reduced. If the metallic oxide contain an unreducible 
 substance, the reduction of the former becomes difficult ; 
 but if a little borax be added the reduction takes place as 
 usual. 
 
 This assay is very easy of execution, and the metal is 
 readily recognised, as by previous assays the nature of it is 
 somewhat ascertained, and the reduction but confirms the 
 previous idea. 
 
 If the metallic oxide be mixed with such a quantity of 
 non-reducible substances that its nature cannot be ascer- 
 tained by previous experiment, how can it be proved that 
 a reducible metal is present ? 
 
 Gahn has solved this question in a very simple manner. 
 ' After having pulverised the substance to be assayed, it 
 is kneaded in the palm of the hand with moistened soda, 
 and the mixture placed on charcoal and exposed to a good 
 reducing flame ; a little more soda is then added, and the 
 blast recommenced. As long as any portion of the sub- 
 
REDUCTION BEFORE THE BLOWPIPE. 217 
 
 stance remains on the charcoal, soda is added in small por- 
 tions, and the blast continued until the charcoal has ab- 
 sorbed the whole of the mass. The first quantities of soda 
 serve to collect the metallic particles scattered in the sub- 
 stance to be assayed, and the final absorption of the latter 
 completes the reduction of any that may remain in the state 
 of oxide. 
 
 ' This done, the burning charcoal is extinguished with 
 a few drops of water ; then having cut out the part which 
 absorbed the soda and assay, grind it to a very fine powder 
 in an agate mortar. This powder is then washed with 
 'water to carry away the finest portion of the charcoal. 
 The grinding and washing are repeated until all the char- 
 coal is washed away. If the substance contained no 
 metallic body, nothing will remain in the mortar after this 
 last washing. But if it contained the smallest quantity of 
 reducible matter, it is found at the bottom of the mortar, 
 as small brilliant plates if it be malleable, or as a fine pow- 
 der if it be brittle or not fusible. In either case the bottom 
 of the mortar is covered by metallic traces, resulting from 
 the friction of the particles of metal against its sides (pro- 
 vided that the quantity of metal contained in the sample 
 be not too small). The flattening of almost imperceptible 
 globules of any malleable metal converts them into shining 
 discs of a perceptible diameter. In this manner may be 
 discovered by the blowpipe, in an assay of ordinary size, 
 less than a half per cent, of tin, and even less than that of 
 copper.' 
 
 The following points in this class of assay ought to be 
 particularly attended to. First, to produce the strongest 
 possible flame, taking care that it covers every part of 
 the assay < Secondly, to leave none of the metal in the 
 charcoal, or lose the smallest quantity in the collection. 
 Thirdly, to well grind the carbonaceous mass. Fourthly, 
 to decant very slowly, so that only the lighter parts may 
 be carried away by the water. Fifthly, not to judge of the 
 result until the whole of the charcoal has been removed, 
 for a small quantity remaining suffices to hide the metallic 
 particles ; and, moreover, the particles of charcoal, viewed 
 
218 BORAX, 
 
 in a certain light, have themselves a metallic lustre, which 
 will deceive an inexperienced eye. Sixthly and lastly, 
 not to trust to the naked eye, however plain the sample 
 may be, but always examine by the aid of a good micro- 
 scope. 
 
 The metals reducible by this process are (besides the 
 noble metals) molybdenum, tungsten, antimony, tellurium, 
 bismuth, tin, lead, thallium, copper, nickel, cobalt, and 
 iron. Amongst these, antimony, bismuth, thallium, and 
 tellurium volatilise easily when they are exposed to a 
 strong heat. Selenium, arsenic, cadmium, zinc, and mer- 
 cury volatilise so completely that they cannot be collected 
 except by means of a small subliming apparatus. 
 
 The reduction can always be effected the first time 
 when the assay contains from 8 to 10 per cent, of metal ; 
 but in proportion as the standard decreases more attention 
 and care must be paid to the washing and recognition of 
 the reduced metal in the mortar. A good system of prac- 
 tice in this experiment is to employ any cupreous sub- 
 stance, and make on it a great number of experiments, 
 taking care to mix it each time with a substance contain- 
 ing no copper ; thus the metallic value will diminish at 
 each new assay, until at last no copper can be found. 
 
 If the substance to be assayed contains several metals, 
 the reduction of their oxides must be made in globo, and 
 a metallic alloy obtained. Some, small in number, are 
 reduced separately. For instance, copper and iron give a 
 regulus of each metal ; copper and zinc, the first gives a 
 regulus of copper, whilst the latter volatilises. But when 
 the result of the operation is an alloy, recourse must be 
 had to the reactions produced by other fluxes to ascertain 
 its constituents. 
 
 BORAX (SODIUM BIBORATE). The borax of commerce 
 must be dissolved in hot water and recrystallised before it 
 can be used in blowpipe analysis. 
 
 Borax may be employed either in crystals, the requisite 
 size for an assay, or in a pulverulent form ; in this case it 
 may be taken up on the moistened point of a knife. Some 
 operators prefer fusing the borax before use, in order to 
 
FUSION WITH BORAX. 210 
 
 V y-> , . "~,-\K\\Pv . 
 
 drive off its water of crystallisation, and thus 
 tumefaction of the crystal on charcoal. 
 
 Borax is employed in the solution or fusion of a variety 
 of substances. It is best to commence by acting upon a 
 scale of the substance to be examined ; because if a pow- 
 der be employed the resulting action cannot be so well 
 ascertained. The following phenomena are to be carefully 
 watched, for in treating any substance with borax it must 
 be particularly noted whether the fusion takes place rapidly 
 or otherwise ; without motion or with effervescence ; if 
 the glass resulting from the fusion is coloured, and if that 
 colour changes in the oxidising or reducing flame ; and, 
 lastly, if the colour diminishes or increases on cooling, and 
 if, under the same circumstances, it loses or retains its 
 transparency. 
 
 Some substances possess the property of forming a 
 limpid glass with borax, which preserves its transparency 
 on cooling, but which, if slightly heated in the exterior 
 (oxidising) flame, becomes opaque and milk-white, or 
 coloured when the flame strikes it in an unequal or inter- 
 mittent manner. The alkaline earths, as yttria, glucina, 
 zirconia ; the oxides of cerium, tantalum, titanium, &c., 
 belong to this class. In order to be certain of this result 
 we must assure ourselves that the glass is saturated to 
 a certain point with either of the above class of bodies. 
 The same thing, however, does not happen with silica, 
 alumina, iron, manganese, &c., oxides, and the presence of 
 silica prevents the production of this phenomenon with the 
 earths ; so that alone they present this peculiar appearance 
 with borax ; but when combined with silica (as natural 
 silicates, for instance) no such effect is produced. This 
 operation has received the name of flaming r , and any 
 substance thus acted upon is said to become opaque by 
 flaming. 
 
 SODIUM AMMONIA-PHOSPHATE (MICROCOSMIC SALT) is obtained 
 by dissolving 16 parts of sal-ammoniac in a very small 
 quantity of boiling water, and mixing with it 100 parts of 
 crystallised sodium phosphate, dissolving the whole with 
 heat, filtering the boiling liquid ; during cooling the double 
 
220 FUSION WITH MICROCOSMIC SALTS. 
 
 salt crystallises. When microcosmic salt is not pure it 
 forms a glass which becomes opaque by cooling. It is then 
 necessary to dissolve it in a small quantity of w r ater and 
 recrystallise it. 
 
 It may be collected in large crystals, or in a pulverulent 
 state. The crystals are in general of a suitable size for 
 ordinary assays. Placed on charcoal, and submitted to the 
 blowpipe flame, it bubbles and swells up, giving off am- 
 monia ; that which remains after this treatment is an acid 
 sodium phosphate, which fuses readily, and forms on cool- 
 ing a transparent and colourless glass. As a reagent, it 
 acts principally by its free phosphoric acid ; and if the salt 
 be employed in preference to the acid, it is because it is 
 less deliquescent, costs less, and passes readily into the 
 charcoal. By means of microcosmic salt we ascertain 
 the action of free acids on any substance we may wish to 
 assay. The excess of acid it contains combines with all 
 bases, and forms a class of double salts, more or less fusible, 
 which are examined as to their transparency and colour. 
 In consequence this flux is used more particularly in the 
 detection of the metallic oxides, most of which impart to 
 it very characteristic colours. This flux exercises on acids 
 a repulsive action. Those which are volatile, sublime ; and 
 those which are fixed remain in the mass, dividing the base 
 with the phosphoric acid, or yielding it up entirely ; in 
 w r hich case they are suspended in the glass without being 
 dissolved. In this respect microcosmic salt is a good test 
 for silicates ; for by its aid silica is liberated, and appears 
 in the glass as a gelatinous mass. 
 
 POTASSIUM NITRATE (NITRE), in long and thin crystals, is 
 employed in hastening the oxidation of those substances 
 which do not readily combine with oxygen in the exterior 
 flame. It is used as follows : The point of a crystal is thrust 
 into the fused bead ; but in order to prevent the cooling 
 of the latter the crystal is held by a pair of pliers, so that 
 when the bead begins to cool it may be withdrawn, the 
 bead reheated, and the crystal employed as before, until 
 the desired effect is produced. 
 
 POTASSIUM BISULPHATE is employed in the detection of 
 
FUSION WITH BISULPHATES. 221 
 
 lithia, boracic acid, nitric acid, hydrofluoric acid, bromine, 
 and iodine. It separates baryta and strontia from the earths 
 and metallic oxides. 
 
 SODIUM BISULPHATE. Professor J. Lawrence Smith* 1 has 
 suggested the use of sodium bisulphate as a substitute for 
 the potassium bisulphate in the decomposition of minerals, 
 especially the aluminous minerals. He finds that ' while 
 the soda salt gives a decomposition at least as complete as 
 the potash salt, the melted mass is very soluble in water, 
 and in the future operations of the analyses there is no 
 embarrassment from a deposit of alum. 
 
 6 The ordinary commercial article is not sufficiently 
 pure for use, and he prepares it from pure sodium carbo- 
 nate, or sodium sulphate that has been purified by recrys- 
 tallisation. In either instance pure sulphuric acid is 
 added in excess to the salt in a large platinum capsule, and 
 heated over a flame, until the melted mass, when taken 
 up on the end of a glass rod, solidifies quite firmly. The 
 mass is then allowed to cool ; moving it over the sides of 
 the capsule will facilitate this operation. When cool it is 
 readily detached from the capsule, is then broken up, and 
 put into a glass stoppered bottle. In almost every instance 
 where we have been in the habit of using potassium bisul- 
 phate the sodium bisulphate can be substituted.' 
 
 VITRIFIED BORACIC ACID is used to ascertain the pre- 
 sence of phosphoric acid and small portions of copper in 
 lead alloys. For quantitative analysis it is generally used 
 to ascertain the quantity of copper contained in a lead ore, 
 and also the amount of copper united with various metals. 
 
 COBALT NITRATE in solution ought to be free from 
 arsenic and nickel, and the solution must be moderately 
 strong. It is used as a test for alumina, magnesia, tin, and 
 zinc, by the blowpipe. 
 
 NICKEL OXALATE is used in qualitative examinations for 
 the detection of potash in a salt which also contains soda 
 and lithia. 
 
 COPPER OXIDE is employed to detect the presence of 
 hydrochloric acid and chlorine. 
 
 * ' American Journal of Science and Arts.' 
 
222 AMMONIUM FLUOEIDE. 
 
 SILICA is, with soda, an excellent test for the presence 
 of sulphuric acid ; and when in combination with borax 
 or soda, separates tin from copper. 
 
 TURNER'S FLUX. This is a mixture of potassium bisul- 
 phate and fluor-spar. It is used for producing coloured 
 flames before the blowpipe. 
 
 AMMONIUM FLUORIDE has been proposed as a blowpipe 
 reagent by Professor N. W. Lord, of the Ohio State Uni- 
 versity. The use of bisulphate of potassium and fluor- 
 . spar as a reagent for developing the flame coloration of 
 boron is well known ; but the alkali present prevents the 
 application of the method for liberating some other bodies 
 in the same way. Fluoride of ammonium, on the contrary, 
 having all the value of fluor-spar as a source of fluorine, 
 admits of much easier application, and is a most useful 
 reagent for detecting the alkalies, boron, and other similar 
 bodies in their mineral combinations. The method of using 
 the reagent is simple. For testing felspar, or similar silicates,. 
 a little of the powdered mineral is mixed with this reagent, 
 then placed on a piece of platinum and moistened with 
 sulphuric acid ; the mixture allowed to stand a few 
 moments, or else gently warmed, taken upon a loop of 
 platinum wire, and tested either in the blowpipe flame or 
 in a Bunsen burner, being dried a little on the wire first. 
 The alkali flame is nearly as well shown as with the pure 
 salts. As the fluoride of ammonium is permanent, is 
 easily obtained free from alkalies and boron, and can be 
 kept indefinitely in a small wooden box, it is always easy 
 to use. 
 
 As a test for boron the reaction is of surprising deli- 
 cacy. The fact that the fluoride of boron is volatile at a 
 temperature far below that required for alkalies, permits 
 thus its detection in borax or any alkaline compound.. 
 To a drop of sulphuric acid placed on a platinum crucible 
 cover, a few grains of the fluoride should be added, and 
 then the mineral (powdered) to form a paste. This is 
 taken as before described on a platinum loop. It should 
 be heated gently until it stops ' sputtering ' from escape 
 of free acid and water (but on no account heated to 
 
CALCIUM FLUORIDE. 223 
 
 redness), then brought not in, but near the lower part of 
 the flame of a Bunsen burner or a good blowpipe flame. 
 A bright green coloration is at once given, untinged by 
 soda-yellow. The coloration is, of course, evanescent, 
 and disappears before the assay is red-hot. A little prac- 
 tice is needed to find the right part of the flame, to get 
 the right heat, and at the same time to draw the volatile 
 boron compound into the heated zone. 
 
 This reaction shows boron very strongly in all speci- 
 mens of tourmaline. With a hand spectroscope the appli- 
 cation of this method gives instant proof of the existence- 
 of boron, potassium, sodium, and lithium, even in very 
 small traces in rocks. 
 
 Borax treated in this way gives a bright green flame,, 
 almost like copper. 
 
 CALCIUM FLUORIDE (FLUOR-SPAR) AND CALCIUM SULPHATE 
 (GYPSUM). These two bodies (deprived of water) are used 
 to indicate the presence of each other. If a small piece 
 of gypsum be ignited in contact with a similar piece of 
 fluor-spar, they soon liquefy at their points of contact ; 
 they then combine, and form, by fusing, a colourless and 
 transparent bead of glass, which becomes enamel white on 
 cooling. Calcium fluoride is thus employed as a test for 
 gypsum, and vice versd. 
 
 Mr. S. D. Poole prefers a mixture of calcium sulphate 
 (selenite) and fluor-spar (in the proportions of two- 
 parts of the former to one of the latter) to Turner's flux 
 described above. It forms an easily fusible bead, which 
 by itself gives only a very faint dull red tint (calcium) to- 
 the flame, but which renders the presence of many elements 
 which give colour most beautifully evident. Thus small 
 portions of lepidolite, petalite, &c., mixed with this flux 
 impart the fine carmine tint of lithium ; copper and stron- 
 tium show their well-known colours, especially after con- 
 tinued blowing. Potassium and sodium minerals (felspar 
 and albite) are at once distinguished. 
 
 This flux is of more limited applicability than that of 
 Turner, because there is no provision in it for the libera- 
 tion of hydrofluoric acid. 
 
224 BONE ASH. 
 
 It serves, when mixed with bisulphate of potash, to 
 detect lithia and boracic acid in their various combina- 
 tions. 
 
 BONE ASHES are employed in the cupel lation of gold and 
 silver. Harkort reduced them to many states of minute 
 division by the processes of sifting and washing. The bones 
 are burned until they become perfectly white, and then 
 freed from any carbonaceous matter that may have ad- 
 hered to them. This being done, they are pulverised in 
 a mortar, and the finer portions separated by a sieve. The 
 remaining powder is then thrown upon a filter, and treated 
 with boiling water, which extracts the soluble matter. The 
 washing, which is then resorted to, is for procuring the 
 bone ashes of a more uniform degree of fineness. The 
 mass from the filter is mixed with water in a cylindrical 
 glass, allowed to settle for a few minutes, and then de- 
 canted ; the coarser powder is deposited at the bottom of 
 the vessel, while the finer passes over suspended in the 
 water. By repeated decantations in this way deposits are 
 obtained of different degrees of fineness ; the last, or that 
 which remains longest floating through the liquid, being 
 the finest. The resulting powders must be kept in sepa- 
 rate bottles. The coarser ashes are used for the cupella- 
 tion of rich silver ores, and the finer for assaying ores 
 in which only a minute quantity of gold or silver is 
 present. 
 
 PROOF LEAD is made use of in cupelling argentiferous or 
 auriferous substances ; it must be free from silver. Dumas 
 states that the best method of obtaining lead in this desir- 
 able state is to decompose the best white-lead by means 
 of charcoal, a it is then impossible for it to contain any 
 other metal. 
 
 TINFOIL is employed to reduce certain peroxides to the 
 state of protoxide. When it is used, a small roll, about a 
 quarter of an inch long, is plunged into the fused button, 
 and heated strongly in the reducing flame : the desired 
 effect is then produced. 
 
 DRY SILVER CHLORIDE. Herr H. Gericke proposes the 
 employment of this compound in qualitative blowpipe 
 
SILVER CHLORIDE IX BLOWPIPE ASSAYS. 225 
 
 assays. In an elaborate paper on this subject, communi- 
 cated to the 'Chemical Gazette' (vol. xiii. p. 189), he 
 says : 
 
 ' Amongst the phenomena which characterise different 
 bodies before the blowpipe, and serve for their distinction, 
 the colour of the flame is of no small importance. This 
 power of colouring the blowpipe flame is not, however, 
 exhibited by all bodies with sufficient intensity to enable 
 them to be distinguished by it with certainty ; and certain 
 substances are consequently usually employed, such as 
 muriatic acid with baryta, strontia and lime, or sulphuric 
 acid, partly to form and partly to set free volatile com- 
 pounds. By this means, however, although the intensity 
 of the coloration is heightened, its duration is not in- 
 creased, as these acids, and particularly muriatic acid, 
 evaporate for the most part before they have acted suffi- 
 ciently, so that the coloration lasts only for a few mo- 
 ments. This defect may be got over by the employment, 
 instead of the volatile muriatic acid, of a chloride, which 
 will retain the chlorine at a high temperature, so that it 
 may only be set free by degrees in small quantities, while 
 the body forming its base may be without action upon the 
 colouring power of the body under investigation. For this 
 purpose chloride of silver appears to be the best, espe- 
 cially as it may readily be prepared in a state of purity. 
 The best plan is to stir it with water into a thick paste, 
 and keep it in a bottle. 
 
 ' In regard to the action of chloride of silver upon the 
 coloration of the blowpipe flame, I have investigated 
 several compounds of potash, soda, lithia, lime, baryta, 
 strontia, copper, molybdenum, arsenic, antimony, and lead, 
 and mixtures of these substances. Chloride of silver, of 
 course, has no action upon borates and phosphates, both 
 of these acids being amongst those which offer the most 
 resistance to the action of heat. 
 
 ' For a support, I employed first of all platinum wire, 
 but this is soon alloyed by the metallic silver which sepa- 
 rates, and thus rendered useless in testing metals. Silver 
 wire is too readily fusible, and also difficult to obtain free 
 
 Q 
 
226 USE OF CHLORIDE OF SILVER 
 
 from copper, which may give rise to errors when in con- 
 tact with chloride of silver. For these reasons, iron wire 
 is best fitted for experiments with chloride of silver, as 
 from its cheapness a new piece may be employed for each 
 experiment, while the silver may readily be obtained in 
 the form of chloride from the broken pieces, If the size 
 of the fragment under examination be sufficient, the plati- 
 num forceps may be employed. 
 
 ' The results at which I arrived, by the employment 
 of chloride of silver, in comparison with those obtained 
 without this reagent, are as follows : 
 
 ' With potash compounds, such as saltpetre, potashes, 
 &c., the flame is decidedly of a darker colour with chloride 
 of silver ; and even in ferrocyanide of potassium, which, 
 when treated by itself with the blowpipe, colours the 
 flame blue, the addition of chloride of silver produces a 
 distinct potash coloration. 
 
 ' The action of chloride of silver upon soda salts is not 
 so favourable ; for although with some, as nitrate of soda, 
 common soda, and labradorite, the flame acquires a more 
 intense yellow colour by the addition of chloride of silver, 
 this reagent produces no observable difference with other 
 soda compounds, such as sulphate of soda and analcime. 
 This also applies to the compounds of lithia, some of 
 which give a finer purple-red colour on the addition of 
 chloride of silver, whilst upon others it has no such effect. 
 
 ' With lime compounds chloride of silver acts favour- 
 ably upon the colouring power. Thus the addition of 
 chloride of silver to calcareous spar or gypsum (in the 
 reduction flame) gives the flame a more distinct yellowish- 
 red colour, but stilbite gives no coloration either with or 
 without chloride of silver. With fluor-spar the coloration 
 cannot well be observed, as this decrepitates too violently 
 under the blowpipe. 
 
 ' The action of chloride of silver upon compounds of 
 baryta and strontia is decidedly advantageous, as both the 
 intensity of the coloration and its duration leave nothing 
 to be desired. Siliceous celestine, which, when heated by 
 itself in the forceps, scarcely coloured the flame, imme- 
 
IN BLOWPIPE ASSAYS. 227 
 
 diately produced a permanent red coloration, when heated 
 with chloride of silver. 
 
 4 Although it appears from the preceding statements, 
 that the employment of chloride of silver presents no 
 advantage with some substances, it may be used with 
 good results in the treatment of mixtures of alkalies and 
 earths. 
 
 4 Thus, with petalite alone the lithia coloration is first 
 produced, and a slight soda coloration is afterwards 
 obtained ; whilst with chloride of silver the soda coloration 
 appears very distinctly after that of the lithia. With 
 lithion-mica alone a very distinct lithia coloration is 
 presented ; but in the presence of chloride of silver a 
 colour is first produced which may lead to the conclusion 
 that potash is present, but the lithia coloration is weak- 
 ened. Ryacolite, heated by itself in the blowpipe flame, 
 only gives a distinct soda coloration ; but with chloride of 
 silver a slight potash coloration is first produced, and the 
 colour of soda then appears very distinctly ; the lime 
 contained in it cannot, however, be detected by the colora- 
 tion of the flame. 
 
 4 Chloride of silver may be employed with still greater 
 advantage with the following metals, but in these cases it 
 is particularly necessary that the operator should become 
 familiar with the colour produced by each individual 
 substance. 
 
 4 With copper compounds, such as red copper ore, 
 malachite, copper pyrites, sulphate of copper, &c., when 
 contained in other minerals so as to be unrecognisable by 
 the eye, the employment of chloride of silver may be of 
 the greatest service, as the smallest quantities of copper, 
 when treated with chloride of silver under the blowpipe, 
 give a continuous and beautiful blue colour to the flame. 
 With chloride of silver the presence of copper may be 
 distinctly ascertained by the blowpipe, even in a solution 
 which is no longer coloured blue by the addition of 
 ammonia. 
 
 4 The employment of chloride of silver will be equally 
 advantageous with molybdenum, as in this case also the 
 
 Q 2 
 
228 USE OF CHLORIDE OF SILVER 
 
 flame gains greatly in intensity. Arsenic, lead, and anti- 
 mony are already sufficiently characterised, the former by 
 its odour, the two latter by their fumes ; but even with 
 these metals chloride of silver may be employed with ad- 
 vantage to render their reactions still more distinct. It is 
 only necessary to observe, that the greenish -blue flame of 
 antimony appears greener and more like that of molyb- 
 denum under the influence of chloride of silver. 
 
 'Chloride of silver may also be employed with com- 
 pounds containing several of the above-mentioned metals. 
 
 ' If bournonite be heated in the oxidation flame of the 
 blowpipe, a fine blue flame is first produced, which in- 
 dicates lead with certainty ; if chloride of silver be now 
 applied, copper is also readily shown. The antimony 
 contained in bournonite cannot be ascertained by the 
 coloration of the flame ; but this may easily be detected 
 upon charcoal, or in a glass tube open at both ends. 
 
 ' Native molybdate of lead, without chloride of silver, 
 only gives a blue colour to the blowpipe flame ; with 
 chloride of silver this blue coloration of the lead comes 
 out more distinctly, but at the same time the tip of the 
 flame, particularly when the reduction flame is employed, 
 appears of a beautiful yellowish-green colour from molyb- 
 denum. 
 
 ' With mixtures of arsenic and copper, or antimony 
 and copper, the flame first acquires a greyish-blue or 
 greenish-blue colour from the oxidation of the arsenic or 
 antimony ; the copper may then be very easily detected 
 by chloride of silver. This applies also to mixtures of 
 arsenic and molybdenum, or antimony and molybdenum ; 
 with .chloride of silver the yellowish-green flame of 
 molybdenum appears distinctly. It will be more difficult 
 to analyse mixtures of arsenic and lead, or antimony and 
 lead, in this manner ; and if a compound contain both 
 arsenic and antimony, these two bodies are not to be 
 distinguished with chloride of silver under the blowpipe. 
 
 ' From these experiments it appears that in blowpipe 
 testing it is advantageous to employ chloride of silver 
 instead of muriatic acid. 
 
IN BLOWPIPE ASSAYS. 229 
 
 ' Chloride of silver is particularly to be recommended 
 in testing metallic alloys for copper. Thus, to test silver 
 for copper, chloride of silver may be applied to the ends 
 of silver wires, and on the application of heat the smallest 
 quantity of copper will furnish the most distinct reaction. 
 This is as sensitive as any of the known copper reactions, 
 and may be performed quickly and easily. In testing 
 metallic alloys for traces of copper, it may be advisable to 
 submit those which contain antimony, zinc, lead, and other 
 volatile metals to roasting^ so as to drive off these metals 
 before the addition of chloride of silver.' 
 
 TINCTURE OF IODINE. Messrs. Wheeler and Ludeking 
 have given, in the ' Transactions of the Royal Society of 
 Canada,' a series of experiments proving that tincture of 
 iodine can be made a very valuable blowpipe reagent. 
 The tincture of iodine is prepared by making a saturated 
 alcoholic solution of the element, which dissolves very 
 readily to a dark red liquid in this menstruum. The 
 usual blowpipe charcoal support is replaced by long thin 
 tablets of plaster-of-Paris, in order to develop the true 
 colours of these varicoloured iodides on a white back- 
 ground. These tablets are prepared by mixing plaster-of- 
 Paris with water to a thin fluid paste, which is poured 
 over a smooth flat surface (as a plate of glass) that has 
 been previously oiled to prevent its adhesion. In a few 
 minutes it will set into a hard cake ; but before this takes 
 place, when it has become stiff, it is divided by a knife or 
 spatula into pieces about 4 by 1-^ inches for use. These 
 tablets are the supports on which these iodide reactions 
 are made by putting the substance on one end, then 
 moistening with the tincture, and blowing with the blue 
 flame, when the volatilised iodides are deposited on its cold 
 surface when suitably inclined. The oxidising flame must 
 be employed in order to prevent the deposition of soot, 
 which tends to interfere by its black film. 
 
 Description of the Iodide Coats. 
 
 Arsenic. A reddish-orange coating. 
 Lead. A chrome-yellow coating. 
 
230 TINCTURE OF IODINE. 
 
 Tin. A brownish- orange coating. 
 
 Silver. A faint greyish : yellow white cold, bright 
 yellow when hot ; close to assay. 
 
 Antimony. An orange-red coating. 
 
 Mercury. A yellow and scarlet, the yellow changing 
 completely to scarlet on standing. 
 
 Selenium. A reddish-brown coating. 
 
 Tellurium. A purplish-brown coating. 
 
 Bismuth. A chocolate brown, fringed with red near 
 the assay. 
 
 Cobalt. A greenish brown, edged with green ; brown 
 coat evanescent, changing into faint green, especially when 
 breathed upon. 
 
 Molybdenum. A deep ultramaririe-blue coating ; close 
 to assay, that is, a permanent oxide. 
 
 Tungsten A faint greenish blue near assay ; that is, 
 a permanent oxide (brought out stronger by dropping 
 more tincture on tablet after the operation). 
 
 Copper. A white coating. 
 
 Cadmium,. A white coating ; becomes bright golden- 
 yellow on blowing ammonium sulphide vapours over it. 
 
 Zinc. A white coating that soon disappears. 
 
 As the copper, cadmium, and zinc iodides are white, 
 the tablets should first be coated with a film of soot 
 (obtained by holding the tablet in a smoky flame) or else 
 charcoal should be used in order to give a black back- 
 ground to the white coats. If to the peculiar velvety 
 chocolate-brown coating of bismuth a drop of dilute 
 ammonia be added, or ammonia vapour be blown over 
 it, the brown disappears, leaving a brilliant red coating. 
 Many of these coatings are more or less evanescent, 
 disappearing on prolonged exposure at ordinary tem- 
 peratures. 
 
 In studying these coats it will be observed that we can 
 now detect tin in the presence of zinc, which has hitherto 
 been impossible with the blowpipe, and that we have a 
 very striking and characteristic reaction for molybdenum. 
 The other coatings are more or less characteristic indivi- 
 dually, and will find favour as confirmatory tests. For 
 
DRY SILVER IODIDE. 231 
 
 mixtures and complicated cases the iodide reactions will 
 not supersede the standard methods. 
 
 DRY SILVER IODIDE. Mr. P. Casamajor has introduced 
 silver iodide into blowpipe work with marked advantages. 
 It gives characteristic sublimates of iodide very beauti- 
 fully and quickly, and it has the advantage over tincture 
 of iodine that it is a dry powder, easily kept in bottles 
 which need not close very perfectly. 
 
 The iodide coatings of mercury, bismuth, and lead are 
 familiar to all chemists who use a blowpipe. They are 
 obtained by V. Kobel's mixture of equal parts of sulphur 
 and of potassium iodide. Silver iodide has over this 
 mixture the advantage that there is no sulphur to give 
 deposits when operations are carried on in glass tubes, 
 and no fumes of sulphur dioxide. It requires less time to 
 obtain the coatings, and they have a more distinct appear- 
 ance. 
 
 In experimenting with silver iodide, mixed with various 
 metallic compounds, the iodide coatings are best deposited 
 in open glass tubes of about 4 inches in length and J inch 
 in diameter. The substance to be tested is mixed into a 
 paste with the silver iodide. A small portion of this 
 mixture is placed at one end of the open tube, and the 
 blowpipe flame is blown on it for a short time. The 
 iodide coatings immediately appear and are seen through 
 the glass. The glass tube may be held by a tongs or 
 simply by a piece of paper as the blowing is not suffi- 
 ciently prolonged to heat the glass tube beyond what 
 paper will stand. 
 
 A small quantity of powdered charcoal or lampblack, 
 mixed with silver iodide and the substance to be tested, 
 gives the characteristic coatings more quickly than dis- 
 tinctly. 
 
 The following metals have given iodide coatings in 
 glass tubes : 
 
 Mercury. In tubes, as on other supports, the yellow 
 and red iodides are produced simultaneously, and streaks 
 of bright red are seen on a yellow ground. 
 
 Bismuth. Yellowish red near the end of the tube and 
 thick brown coating beyond. 
 
DRY SILVER IODIDE. 
 
 Lead and Tin. Both these metals give bright yellow 
 deposits which retain their colour when cold. These de- 
 posits cannot be distinguished one from the other, both 
 being equally bright. In the case of tin a very strong 
 smell of iodine is given off, which is possibly due to stannic 
 iodide. 
 
 Arsenic. Near the end of the tube to which the flame 
 is applied there is a yellow deposit ; beyond this a white 
 coating of arsenious acid. The yellow portion turns white 
 on cooling, but becomes yellow again when the tube is 
 heated over a flame. 
 
 Antimony. The orange-red coating given by this metal 
 becomes quite faint on cooling, but the colour becomes 
 bright again when the tube is heated. 
 
 Zinc. The deposit is white both when cold and when 
 hot ; the fumes are not very abundant, much less so than 
 those due to lead or tin. 
 
 Iron makes a deposit which may be considered as 
 characteristic from the fact that the yellow coating in the 
 tube is streaked with distinct dashes of brown. The 
 yellow portion becomes white on cooling, but the brown 
 streaks do not change. 
 
 Thallium. A yellow coating is deposited as with most 
 metals. After this has taken place, if a reducing flame 
 touches the deposit, this fades, leaving a grey tinge with 
 an edge of purple. This seems to be characteristic of 
 thallium. 
 
 Ccesium, Rubidium, and Lithium have not given de- 
 posits which can be called characteristic. The deposit 
 from caesium differs from those of the other two metals in 
 being less volatile. The caesium deposit does not extend 
 far beyond the heated end of the tube. By increasing the 
 heat it melts, but does not move forward. 
 
 Chromium gives a white coating which remains at the 
 hot end of the tube. The portion nearest to this end by 
 further heating becomes of a pale reddish brown. 
 
 Manganese. Yellow hot, but white when cold, like 
 deposits from many other metals. 
 
 Molybdenum. Beyond the yellow coating, which turns 
 
SODA-PAPER. 
 
 233 
 
 white on cooling, are distinct blue streaks, which are very 
 characteristic. These were first observed by Messrs. 
 Wheeler and Ludeking, by treatment with tincture of 
 iodine on tablets of plaster-of-Paris. With a glass tube, the 
 blue streaks extend through the whole length of the tube. 
 
 Manganese and Uranium give deposits which are 
 yellow when hot, and white when cold. These are too 
 common to be characteristic. 
 
 Deposits have been obtained on charcoal and on thin 
 sheets of iron, either on the metallic surface or on a coat 
 of soot, by the use of silver iodide and metallic compounds. 
 Some of these deposits are very good, but they are not so 
 uniform for the same metal as deposits obtained in glass 
 tubes. 
 
 SODA-PAPER. Mr. Forbes writes as follows in the 
 ' Chemical News ' : ' As it would be impossible to submit 
 any powdered substance to the direct action of the blow- 
 pipe flame without its suffering mechanical loss, some 
 means must be employed for holding the particles together 
 until they are so agglutinated by the heat that no such 
 loss need be apprehended ; this is secured by the use of 
 the soda-paper envelope or cornet, as devised by Harkort. 
 For this purpose slips of thin slightly sized writing paper, 
 about 1 J inch long by 1 
 inch broad, are steeped in a 
 solution of one part crystal- 
 lised pure carbonate of soda 
 (free from sulphate) in two 
 parts of water. When dried 
 these are used for forming 
 small cylindrical cornets, by 
 rolling them round the ivory 
 cylinder, fig. 73 d, previously 
 described. A bottom is formed 
 to them by folding down a 
 portion of their length on to 
 the end of the cylinder, which 
 is then pressed firmly into the corresponding mould in the 
 blowpipe anvil, and which, upon the withdrawal of the 
 
 F IG . 73. 
 
234 SODA-PAPER. 
 
 cylinder, serves as a support until they are filled with the 
 assay from the scoop in which the assay and flux have 
 been mixed. After pressing the assay down, the super- 
 fluous paper is cut off, leaving only sufficient when folded 
 down upon the contents of the cornet, to form a paper 
 cover to the top similar to the hollow of the cornet. The 
 assay is then ready for placing in a bore in the charcoal, 
 formed by the charcoal borer c, fig. 71, and is then 
 submitted to a reducing fusion.' 
 
 GENERAL ROUTINE OF BLOWPIPE OPERATIONS. 
 
 Size of the Assay. The morsel operated on is suffi- 
 ciently large when the effect of the heat and the fluxes added 
 can be distinctly discerned. The size of the assay-piece 
 generally recommended is much too large ; its size ought 
 to be about that of a mustard seed ; that of the flux added, 
 about the size of a hemp-seed. It should in general be 
 previously reduced to fine powder. 
 
 When a large piece is employed, the experiment con- 
 sumes much more time and requires much more labour 
 than a smaller piece. It is only in reductions that a 
 larger piece may be successfully employed, because in 
 that case the more metal produced, the more readily can 
 its nature be ascertained. Having thus endeavoured to 
 fix the size of the assay, we will now lead our readers to 
 the operations necessary in blowpipe analysis, and in the 
 order in which they are to be performed. 
 
 First. The substance is heated in the closed tube, or 
 mattrass, over a spirit lamp. It may, by this treatment, 
 decrepitate or give off water, or some other volatile sub- 
 stance. 
 
 Secondly. It is heated gently on charcoal, by aid of 
 the blowpipe ; and, as soon as warm, withdrawn from the 
 heat, and the odour given off ascertained : volatile acids, 
 .arsenic, selenium, or sulphur, may be present. The odour 
 thus produced by the oxidising flame must be compared 
 with that produced by the reducing flame ; if there is any 
 difference, it must be carefully noted. Sulphur, selenium, 
 
OPERATIONS IN BLOWPIPE ANALYSIS. 235 
 
 <&c., are best detected in the oxidising flame, and arsenic 
 in the reducing flame. 
 
 Thirdly. The substance is examined as regards its 
 fusibility. If it be in grains, it is better acted upon on 
 charcoal, notwithstanding its liability to escape on the first 
 insufflation, when not very fusible. But if we can choose 
 the form, it is better to knock off a small splinter, by means 
 of the hammer, and hold it in the flame by the platinum- 
 pointed pincers. A fragment with the most pointed and 
 the thinnest edges ought to be selected. By thus acting, 
 we can always ascertain at a glance if the substance be 
 fusible or not. Infusible substances retain their sharp 
 points and angles, which can be ascertained immediately by 
 means of a microscope. The points are merely rounded in 
 bodies of difficult fusibility, and in substances of easy 
 fusion are rendered globular. 
 
 Certain substances, and particularly some minerals, 
 change both aspect and form when exposed to the 
 blowpipe flame, without entering into fusion ; some swell 
 up like borax ; some of them fuse after tumefaction ; 
 others keep in that state without fusion. Some minerals 
 give off a sort of foam on fusing, giving rise to a kind of 
 blebby glass, which, although transparent itself, does not 
 appear so, on account of the multitude of air-bubbles it 
 contains. 
 
 This bubbling and tumefaction takes place in the 
 greater part of the minerals only at that temperature at 
 which all the water is disengaged ; and these ramifications 
 appear to proceed from a new molecular arrangement, 
 produced by the action of heat on the constituent parts of 
 the substance. It cannot be said that the expansion of a 
 particular part of the substance or its formation into gas, 
 gives rise to this, because it most often happens in those 
 substances which contain no such substance. The minerals 
 which generally give these indications are the double 
 silicates of lime or alkali, and alumina. It sometimes 
 disappears after a few instants, and occasionally lasts as 
 long as the substance is kept in fusion. In the latter 
 case, it seems that the assay takes carbonic acid from the 
 
236 PRELIMINARY BLOWPIPE OPERATIONS. 
 
 flame, which carbonic acid is transformed by the charcoal 
 into carbonic oxide, and it is that gas which causes the 
 bubbles. 
 
 In the employment of fluxes, it is necessary to continue 
 the blast for a sufficiently long time, because some sub- 
 stances appear infusible at the commencement of the oper- 
 ation, and gradually yield to the influence of the flux, and 
 in about two minutes enter into full fusion. The substance 
 is best added in small quantities, and no new dose must be 
 introduced until the former one is acted upon, so that at 
 last the glass arrives at that degree of saturation that it can 
 dissolve no more : it is at this particular point that the 
 reactions are most vivid and plain. Beads of glass, not so 
 saturated, do not give such decided indications. 
 
 Occasionally, in operating with a flux in the reducing 
 flame, it happens that the assay-bead reoxidises during the 
 cooling of the charcoal, and thus the labour of a preceding 
 operation is lost. In order to obviate this inconvenience, 
 the charcoal is turned over, so that the bead may fall in a 
 yet liquid state on some cold body, as a porcelain plate. 
 
 When the colour of the bead is so intense that it ap- 
 pears opaque, its transparency can be proved by holding 
 it opposite to the flame of a lamp ; the reversed image of 
 the flame can then be seen in the bead, tinged with the 
 colour imparted to the flux by the body under experiment. 
 The globule may also be flattened by a pair of pliers before 
 it cools, or it may be drawn into a thin thread. In either 
 of the last-mentioned cases its colour can readily be ascer- 
 tained. 
 
 Minerals exposed to the exterior and interior flame, 
 either with or without fluxes, present a variety of pheno- 
 mena, which ought to be carefully noted, and which, 
 collectively, form the result of the assay. The smallest 
 circumstance must not be overlooked, because it may lead 
 us to ascertain the presence of a substance not suspected. 
 It is necessary, in all cases, to make two assays, and com- 
 pare the separate results ; because it sometimes happens 
 that an apparently trivial fact had been overlooked in the 
 first series of operations, which materially conduces to the 
 good result of the experiment. 
 
DISCRIMINATION OF MINERALS. 237 
 
 DISCRIMINATION OF MINERALS.* 
 
 Selecting those minerals which are remarkable for 
 their wide distribution, and at the same time those which 
 are of value, whether rare or frequent, the number which 
 it is advisable for the explorer to know is reduced to about 
 seventy-five, and it is only to this limited number that 
 reference will be made in this chapter. 
 
 The distinctions in minerals lie in their chemical and 
 physical characters ; but as the appearance of a mineral 
 may afford little or no indication of what these are, they 
 have to be discovered by testing. 
 
 These characters will be described in detail, and the 
 modes of testing will be indicated, where necessary. 
 
 The characters include crystalline form, mode of 
 fracture, colour, lustre, transparency, taste, odour, feel, 
 hardness, specific gravity, fusibility, and chemical com- 
 position. 
 
 Crystalline Form. 
 
 The crystalline form is an important guide in many 
 cases, but crystals are subject to so many modifications, 
 that this character ceases to be a practical one except to 
 those persons who are familiar with its geometrical laws. 
 Stil] it is very useful to know a few common primary 
 forms such as the following : 
 
 CUBE. 6 sides or faces, each square (fig. 74). 
 
 OCTAHEDRON. 8 sides, each an equal-sided triangle 
 (fig. 75). 
 
 DODECAHEDRON. 12 sides, each rhombic, (shaped like 
 the diamond of a pack of cards) (fig. 76). 
 
 TETRAHEDRON. 4 sides, each triangular (fig. 77). 
 
 * This method of distinguishing the various minerals, ores, etc. which are 
 likely to be met with during an exploration is partly condensed from a very 
 valuable pamphlet by Alexander M. Thompson, D.Sc., Professor of Geology at 
 the University of Sydney, entitled ' Guide to Mineral Explorers in Distin- 
 guishing Minerals, Ores, and Gems.' The instructions given will not be found 
 to involve greater difficulties than can be overcome by a little practice and 
 perseverance. The appliances which are needed may be bought for a few 
 shillings, and will be found too trifling in weight and bulk to incommode the 
 traveller. 
 
238 
 
 DISCRIMINATION OF MINERALS. 
 
 EHOMBOHEDRON. 6 sides, each rhombic (fig. 78). 
 PRISM. Any column with three or more sides ; when 
 
 FIG. 74. 
 
 FIG. 75. 
 
 FIG. 76. 
 
 FIG. 77. 
 
 FIG. 80. 
 
 FIG. 79. 
 
 Fm. 81. 
 
 placed on its base, it may stand straight or oblique ; it may 
 terminate abruptly with a flat face, or come off to a point, 
 blunt or sharp, like a pyramid (figs. 79, 80, 81). 
 
 Mode of Fracture. 
 
 The mode in which a mineral breaks when smartly 
 struck with a hammer, or pressed with the point of a knife, 
 is a character of importance. Many can only be broken 
 in certain directions ; for instance, a crystal of calc-spar 
 can only be split parallel to the faces of a rhombohedron ; 
 many crystals break more easily in one direction than in 
 
DISCRIMINATION OF MINERALS. 239 
 
 others. Whenever a mineral breaks with a smooth, flat, 
 even surface, it is said to exhibit cleavage. Cleavage 
 always depends upon the crystalline form. But minerals 
 often break in irregular directions, having no connection 
 whatever with the crystalline form, and this kind of 
 breaking is called fracture ; the broken surfaces are usually 
 irregular or conchoidal (i.e. with concave and convex out- 
 lines, like shells). 
 
 Lustre. 
 
 Some minerals have a brilliant lustre like that of 
 metals ; in others the lustre resembles that of glass, silk, 
 resin, or wax ; while others are dull, or destitute of lustre. 
 The lustre of the diamond is called adamantine. 
 
 Colour and Streak. 
 
 Minerals may be colourless, white, black, or of any 
 colour, either dull or brilliant. The same mineral may 
 present a variety of similar tints, or even distinct colours. 
 It often happens that a mineral, which when viewed in a 
 solid mass possesses a distinct colour, affords a powder 
 which has a colour different from that of the solid mass, 
 or is even destitute of colour, that is to say, white or 
 nearly so. The dust formed on scratching a mineral with 
 a knife, or by a splinter of quartz, or by a diamond, is 
 termed the ' streak ' : it has usually the same colour as the 
 powder. The colour of a mineral and its streak may cor- 
 respond ; or the mineral and its streak may possess diffe- 
 rent colours ; or the mineral may be coloured, while its 
 streak is colourless. 
 
 For instance, cinnabar has both a red colour and a red 
 streak. 
 
 Specular iron has a black colour, but a red streak. 
 
 Sapphire has a blue colour, but a white or colourless 
 streak. 
 
 The streak of most minerals is dull and pulverulent, 
 but a few minerals exhibit a shining streak, like that 
 formed on scratching a piece of lead or copper. This 
 kind of streak is distinguished by the name of Metallic. 
 
240 DISCRIMINATION OF MINERALS. 
 
 In judging the colour of a mineral a surface quite free 
 from tarnish should be chosen. 
 
 The colour of a transparent gem can be seen to best 
 advantage by immersing it in water, about half an inch 
 below the surface. 
 
 Hardness. 
 
 This character is of great importance in distinguishing 
 minerals ; it implies the degree of facility with which the 
 particles may be separated by cutting or scratching. The 
 diamond is the hardest substance known, as it will scratch 
 all others. Talc is one of the softest minerals. Other 
 minerals possess intermediate degrees of hardness. To 
 express how hard any mineral is, it becomes necessary to 
 compare it with some known standard. Ten standards of 
 different degrees have been chosen, and are given in order 
 in the following scale : 
 
 1. Talc. 6. Felspar. 
 
 2. Gypsum. 7. Quartz. 
 
 3. Calc Spar. 8. Topaz, 
 
 4. Fluor Spar. 9. Corundum, or Sapphire. 
 
 5. Apatite. 10. Diamond. 
 
 The hardness of a mineral may often be found by 
 drawing the point of a steel knife across it. For instance, 
 the slightest pressure will suffice to scratch talc ; fluor-spar 
 is not so easily scratched as calcite ; the greatest pressure 
 is needed to scratch felspar ; and quartz does not yield to 
 the knife at all. 
 
 The hardness may also be found by scratching one 
 mineral with another ; thus, the diamond will scratch all 
 other minerals ; corundum scratches topaz, topaz scratches 
 quartz, quartz scratches felspar ; and so on. 
 
 If on drawing a knife across a mineral it is impressed 
 as easily as calcite, its hardness is said to be 3. If a 
 mineral scratches quartz, but is itself scratched by topaz, 
 its hardness is between 7 and 8. 
 
 In trying the hardness of a mineral a little judgment 
 is necessary ; for instance, a sound portion of the mineral 
 
DISCRIMINATION OF MINERALS. 241 
 
 must be chosen ; a sharp angle used in trying to scratch : 
 a streak of dust on scratching one mineral with another 
 may come from the waste of either, and it cannot be 
 determined which is the softer, until after wiping off the 
 dust, and viewing with a lens. 
 
 The use of the above scale, however, implies an ac- 
 quaintance with the ten standard minerals. It is very 
 desirable to have specimens of these minerals for reference. 
 But a collection of the following common minerals, which 
 can easily be procured, will be found quite enough for 
 ordinary purposes viz. : calc-spar, felspar, quartz, topaz 
 (white or yellow), and corundum or sapphire. 
 
 By the test of hardness clear distinctions may be drawn 
 between minerals which resemble each other ; for instance, 
 iron pyrites and copper pyrites are similar in appearance, 
 but copper pyrites can easily be scratched with a knife, 
 whereas iron pyrites is nearly as hard as quartz and cannot 
 be impressed with a knife at all. 
 
 In determining the hardness of small stones, it is most 
 convenient to fasten them by heating upon a stick of seal- 
 ing wax. 
 
 Specific Gravity. 
 
 By specific gravity is meant the comparative weight of 
 equal bulks. Water is taken as the standard of compari- 
 son ; the specific gravity of a mineral is a number showing 
 how many times it is, bulk for bulk, heavier than water. 
 
 The specific gravity of water is called 1, of gold 19, 
 implying that, if equal bulks of gold and water were taken, 
 the gold would weigh 19 times as heavy as the water. The 
 specific gravity of a mineral can be found by weighing it 
 first in air in the usual manner, and then observing how 
 much of its weight it loses, when suspended from the arm 
 or pan of a balance, and allowed to hang freely in water. 
 If a piece of quartz weighing 26 grains is attached by a 
 hair or thin cotton to the scales, and weighed whilst hang- 
 ing in water, it will be found to weigh only 16 grains ; it 
 thus loses 10 grains, or if of its entire weight. Similarly 
 gold would lose -fa of its weight. 
 
 R 
 
242 DISCRIMINATION OF MINERALS. 
 
 Minerals differ very widely in the proportion of weight 
 which they lose in water, but the same mineral invariably 
 loses the same proportion ; for instance : 
 
 Quartz loses -J-g- of its weight ; topaz ^g- ; sapphire ^ ; 
 zircon i-g. ; tin ore f [}. 
 
 These proportions depend upon the specific gravity of 
 these minerals. The specific gravity of water is called 1, 
 .of quartz 2*6, of topaz 3*5, of sapphire 4*0, of gold 19. 
 
 In estimating how much weight a mineral loses in 
 water, a very delicate balance is required when the weight 
 in air is under 10 grains ; but for portions weighing heavier 
 than this, a common balance turning readily to a grain 
 may be used for practical purposes. The mineral must be 
 sound throughout, and free from any pores or cracks, and 
 its surface should be rubbed over with water, before im- 
 mersing it, to prevent bubbles of air adhering, which would 
 falsify the result. In careful trials distilled or rain water 
 should be used. A trial .of specific gravity can have no 
 value, unless it is made on a pure portion of a mineral, quite 
 free from any adhering foreign matter. 
 
 A rough estimate of specific gravity can be formed from 
 the feeling of pressure in shaking any mass loosely in the 
 palm of the hand : in this way it can be judged whether 
 the specific gravity is high or low. 
 
 EULE. The rule for finding the specific gravity is to 
 divide the weight of the mineral in air by its loss of weight 
 in water. Example: A piece of quartz weighed 1,398 
 grains in air, and 862 grains in water ; here the loss of 
 weight is 536, and the weight in air divided by this number 
 is 2 -6, which is the specific gravity of quartz. 
 
 For separating from each other minerals of different 
 specific gravities R. Breon proposes the use of melted mix- 
 tures of lead and zinc chlorides in different proportions, so 
 as to yield liquids ranging from 2' 4 to 5*0 in specific 
 gravity. 
 
 Sonstadt* has devised a very ingenious and useful 
 method of taking specific gravities of minerals, when a 
 rough and ready method of discriminating between bodies 
 
 * ' Chemical News,' March 20, 1874, vol. xxix. p. 127. 
 
DISCRIMINATION OP MINERALS. 243 
 
 of similar appearance but of different specific gravities is 
 all that is needed. He takes a solution of potassium 
 iodide saturated at the common temperature, and dissolves 
 in it as much iodide of mercury as possible. It will 
 then dissolve more potassium iodide, then more mercuric 
 iodide, and so forth. The iodides dissolve very slowly at 
 the last, and as it is best not to accelerate the solution by 
 the application of heat, considerable time must be allowed 
 when a liquid of maximum strength is required. The 
 solution, after filtering, is fit for use, and it may be 
 obtained without danger of its crystallising above zero, 
 of sp. gr. 3*085 at 12 C. This liquid is transparent, very 
 mobile, filters easily', is of an amber colour, gives no preci- 
 pitate on addition of water, and does not readily lose or gain 
 water on long exposure to the atmosphere. The solution 
 undergoes no perceptible chemical change by free exposure 
 to the air for many months. It may be diluted to any 
 extent, and then concentrated by heat without injury. 
 
 When this liquid is diluted so that quartz will just 
 float in it, any mineral heavier than quartz will of course 
 sink. As many minerals closely resemble quartz in external 
 characters, and can be distinguished more readily by their 
 specific gravity than in any other way, this method of 
 testing the specific gravity furnishes a very quick and 
 certain means of discrimination. A small bottle of the 
 liquid, containing say half an ounce, and carried in the 
 waistcoat pocket, will enable the mineralogist to make a 
 great many examinations in his walks as to whether the 
 specimens examined are above or below a certain specific 
 gravity. The smallest distinctly visible particle of a 
 mineral is enough for an estimation. After the bottle 
 becomes clogged with specimens, the liquid may be poured 
 out for further use, and if the washings are added and the 
 whole concentrated there need be no sensible waste. 
 
 Mr. Hardman suggests that with this liquid it would 
 be possible to separate completely the three constituent 
 minerals of granite mica, felspar, and quartz weigh 
 them, and estimate absolutely their percentage. 
 
 Professor Church has used Sonstadt's solution exten- 
 
 R 2 
 
244 DISCRIMINATION OF MINERALS. 
 
 sively for years in the separation of rare minerals from 
 their siliceous gangue. He says he has often identified 
 the minerals of a crushed fragment of rock by introducing 
 a drop of the solution into the ' live box ' under the 
 microscope. By using solutions of increasing strength, or 
 adding the solid salt, the felspars, the quartz, and then 
 some of the heavier minerals may be successively brought 
 to the surface and thus into focus. 
 
 Fusibility. 
 
 Some minerals can be easily fused ; others only with 
 difficulty, while others resist the highest heat which can 
 be applied to them. There are such wide differences 
 between the various degrees of fusibility of minerals, that 
 this character helps greatly in distinguishing them. The 
 fusibility is most readily tested by holding a small splinter 
 of the mineral with a forceps in a candle flame, urged 
 by the blowpipe ; or the mineral may be laid upon a piece 
 of charcoal, and the flame directed upon it by the blow- 
 pipe. Some minerals fly to pieces when heated, others 
 swell up, or give off peculiar and characteristic odours. 
 
 Chemical Characters. 
 
 The chemical composition of a mineral is one of the 
 most important facts to know respecting it; the various 
 processes for estimating this form a science by them- 
 selves ; but only the tests depending upon the use of the 
 blowpipe will be described here. A great deal can be 
 learnt respecting a mineral by a few simple trials with the 
 blowpipe. The only requirements are a common blow- 
 pipe, a candle, a forceps or pliers, a piece of No. 22 plati- 
 num wire, two inches long, dried sodium carbonate, dried 
 borax, and potassium cyanide. The charcoal selected for 
 these experiments should be free from cracks and openings. 
 By dry sodium carbonate is meant not merely dry to the 
 touch, but quite free from water ; this may be prepared 
 from recrystallised common washing soda, by expelling 
 the water which it contains ; the washing soda is put in a 
 shallow clean iron dish, and placed over a clear fire until 
 a white dry powder is formed ; too strong a heat might 
 
DISCRIMINATION OF MINERALS. 245 
 
 fuse the dry powder. A quarter of an ounce may be kept 
 in a well-corked bottle or tube for use. Sodium bicarbon- 
 ate may be used instead without previous heating ; or if 
 the bicarbonate be moderately heated, it loses weight, and 
 becomes sodium carbonate, quite free from water like the 
 above. 
 
 The borax is to be dried in the same way ; a quarter 
 of an ounce will be enough ; it is convenient to keep the 
 platinum wire in the same tube. Unless these tubes are 
 well corked these chemicals reabsorb moisture. For test- 
 ing tin ore it is useful to have a little potassium cyanide 
 kept in a bottle with the cork and rim well covered 
 with melted bees'-wax ; it would otherwise liquefy, by 
 -absorption of moisture, and become useless. It is a 
 most dangerous poison, and the greatest caution must 
 be observed in its use. 
 
 Three kinds of effect can be produced by the blowpipe : 
 first, it can be used as a source of heat for testing fusibility ; 
 secondly, it can add oxygen to a mineral ; thirdly, it can 
 take oxygen away. 
 
 The three principal means of chemically testing 
 minerals before the blowpipe are (1) with borax; (2) on 
 charcoal, usually with the addition of sodium carbonate ; 
 (3) by holding in the point of oxidation. 
 '. EXPERIMENT No. 1. Many metals impart a colour to 
 fused borax, by which their presence can be recognised. 
 To try this experiment a bead of fused borax must first 
 be obtained on the platinum wire. The end of the wire 
 is bent into a loop or ring about the twelfth part of an 
 inch in diameter. The wire is then heated in the blow- 
 pipe flame, and dipped whilst hot into the borax : the por- 
 tion of borax that adheres is then fused on to the wire in 
 the blowpipe flame, and the hot wire is again dipped ; this 
 is repeated until the loop contains a glass-like bead of fused 
 borax. If the bead has become cloudy, the soot causing 
 this may be burned off in the oxidising point of the flame. 
 Having thus obtained a clear, colourless, transparent bead, 
 -the next step is to add to it a minute portion of the mineral 
 which is to be tested. By touching a little of the finely 
 
240 DISCRIMINATION OF MINERALS. 
 
 pulverised mineral with the borax bead, while softened by 
 heat, enough will adhere to the bead for a first trial. The 
 bead is then kept at a white heat in the oxidising point of 
 the flame for a few seconds, and on removal its colour is 
 noted, both whilst hot and when cold. If no colour is 
 imparted a fresh trial may be made with a larger quantity 
 of the powder ; but if the bead is opaque owing to the 
 depth of colour, as is often the case, a fresh experiment 
 must be made, using a still smaller quantity of the powder. 
 The colour can only be fairly judged in a perfectly trans- 
 parent bead. If no colour can be obtained in the oxidising 
 flame, further experiment with the borax bead is needless ; 
 but if a colour is obtained, it is then advisable to try the 
 effect of the reducing flame upon the same bead. The 
 following observations and inferences may result from this 
 test : 
 
 Colour of Bead 
 
 in 
 
 Showing the 
 
 OXIDISING FLAME. EEDUCING FLAME. 
 
 presence of 
 
 Green (hot) ; Blue (cold). 
 
 Red. 
 
 COPPER. 
 
 Blue (hot and cold). 
 
 Blue. 
 
 COBALT. 
 
 Amethyst. 
 
 Colourless 
 
 MANGANESE. 
 
 Green. 
 
 Green. 
 
 CHROMIUM. 
 
 Bed or Yellow (hot). 1 
 Yellow or colourless (cold). f 
 
 Bottle Green. 
 
 IRON. 
 
 Violet (hot) ; Ked-Brown (cold). 
 
 Grey and Turbid, 
 
 NICKEL. 
 
 
 difficult to ob- 
 
 
 tain. 
 
 This mode of testing may often be used to prove the 
 presence of the above-mentioned metals. 
 
 It requires some practice before trustworthy results 
 can be obtained in reducing ; the reduced bead, if brought 
 out of the flame at a white heat into the air, may at once 
 oxidise ; but this may be prevented by placing it inside 
 the dark inner cone of an ordinary candle flame, and 
 allowing it to cool partially there. 
 
 EXPERIMENT No. 2. The mode of testing with sodium 
 carbonate on charcoal is performed as follows : A sound 
 piece of charcoal half an inch square is chosen, and a neat 
 cavity is scooped out on its surface, into which is placed 
 a mixture containing the pulverised mineral to be tested, 
 with three or four parts of sodium carbonate, the whole 
 not exceeding the bulk of a pea ; after lightly pressing 
 
 
DISCRIMINATION OF MINERALS. 247 
 
 the mixture into the cavity, the blowpipe flame may be 
 cautiously applied to it ; and afterwards, when the mix- 
 ture no longer shows a tendency to fly off, the charcoal 
 may be advanced nearer to the blowpipe, and finally be 
 kept at as high a temperature as possible in the reducing 
 part of the flame. 
 
 In testing for tin ore, a piece of potassium cyanide 
 about the size of a pea may be placed upon the mixture 
 after the first application of heat, and the further applica- 
 tion of heat may then be continued. 
 
 This treatment is designed to extract metals from 
 minerals ; it favours in the highest degree the removal of 
 oxygen. But like the borax test, jt is limited in its appli- 
 cation ; it can only be used to detect certain metals. The 
 failure of the test, in any case, must not be looked upon as 
 a conclusive proof of the absence of the particular metal 
 sought : for instance, copper can be easily abstracted from 
 copper carbonate by this test, but not from copper pyrites. 
 Still the test is a most valuable and indispensable one to 
 the mineralogist. 
 
 The test is complete when the metal is obtained as 
 a globule, in the cavity of the charcoal ; in many cases 
 the globule will be found surrounded with the oxide 
 of the metal, forming an incrustation on the charcoal ; 
 and the colour of such incrustation should be carefully 
 noted, both at the moment of removal, from the flame, 
 and after cooling. By pressing the globule between 
 smooth and hard surfaces, it can be seen whether the 
 metal is flattened out (or malleable), or crushed to pieces 
 (brittle). 
 
 The following observations and inferences may result 
 from this test: 
 
 GLOBULE. INCRUSTATION. Presence of 
 
 Yellow (malleable). None. GOLD. 
 
 White do. Do. SILVER. 
 
 Ked do. Do. COPPER. 
 
 White do. White. TIN. 
 
 Do. do. Bed (hot) ; Yellow (cold). LEAD. 
 
 Do. (brittle). Do. do. BISMUTH. 
 
 None. Yellow (hot) ; White (cold). ZINC. 
 
 White, brittle, giving off White. ANTIMONY. 
 fumes when removed 
 from the flame. 
 
248 DISCRIMINATION OF MINERALS. 
 
 EXPERIMENT No. 3.- In addition to these substances 
 there are others which occur abundantly in minerals, and 
 which may be recognised by the blowpipe with the greatest 
 ease ; for instance, sulphur and arsenic. These may be 
 discovered by heating a fragment of the mineral, supported 
 on a piece of charcoal, or held in a forceps, in the oxidising 
 point of the flame, and noticing the odour which is given 
 off; a smell of burning sulphur indicates that the mineral 
 contains that substance, and white fumes having a garlic 
 odour indicate the presence of arsenic. 
 
 Mercury, antimony, and other substances may escape 
 as fumes when heated in this manner. 
 
 Requirements for Testing Minerals. 
 
 The requirements already mentioned are blowpipe, 
 candle, forceps, platinum wire, dried borax, dried sodium 
 carbonate, potassium cyanide, charcoal ; also the minerals 
 for comparisons of hardness namely, calc-spar, felspar, 
 quartz, topaz, sapphire. In travelling it is well to dis- 
 pense with the grain scales and weights for taking specific 
 gravities. It would be dangerous to attempt to travel 
 with nitric acid. In addition to these it will be found 
 useful to have a steel pocket-knife, one blade of which 
 should be kept magnetised, which may be easily done by 
 touching it occasionally with a strong magnet ; a small 
 iron spoon for heating minerals, such as cinnabar, over a 
 candle flame ; also a small pocket lens. 
 
 DESCRIPTION OF MINERALS. 
 QUARTZ AND THE SILICATES. 
 
 Perhaps no other mineral presents such a great variety 
 of forms and colours as quartz, and no mineral occurs in 
 greater abundance. When pure it consists of silica only, 
 but it is usually contaminated with other ingredients, 
 principally alumina, iron, and clay. The impure varieties 
 of quartz compose most of the pebbles and sand of the 
 soil. 
 
DISCRIMINATION OF MINERALS. 249 
 
 Out, of the six hundred minerals at present known, at 
 least two hundred and fifty contain silica in greater or less 
 proportion, and are hence termed silicates. 
 
 Silicates are indispensable in the manufacture of glass, 
 porcelain, earthenware, and for other purposes ; but they 
 exist in such profusion that their economic value is ex- 
 ceedingly trifling. The great majority of the silicates are 
 purely objects of scientific interest ; a few forms of silica 
 &re esteemed as gems, such as the precious opal, and some 
 varieties of quartz ; also the silicates, topaz, emerald, 
 zircon or hyacinth, garnet and carbuncle, the tourmaline, 
 .and some others. Handsome varieties of serpentine are 
 often used in ornamental stonework. 
 
 1. QUARTZ. 
 
 Chemical Composition. Silica. 
 
 Crystallised in six-sized prisms, terminated by pyramids. 
 Sides of the crystal often marked across with fine parallel 
 lines. Transparent or opaque. Colourless, or of various 
 colours. Glassy lustre. Fracture irregular, conchoidal. 
 Specific gravity, 2'6. Hardness, 7. Cannot be scratched 
 with a knife ; scratched by topaz, zircon, sapphire, and 
 diamond, and thus easily distinguished from these gems. 
 Quartz scratches glass with facility ; felspar and many 
 other minerals can be scratched by quartz. The irregular 
 fracture, the fine parallel markings, and the hardness gene- 
 rally suffice to distinguish quartz. 
 
 Infusible in the blowpipe flame. 
 
 The following are the chief varieties of quartz ; the 
 differences are due either to their mode of formation or 
 the presence of impurities. They have the same general 
 characters as pure quartz, 
 
 1. TRANSPARENT VARIETIES : 
 
 BOCK CRYSTAL. Pure, transparent, colourless quartz. 
 Used for spectacle glasses and ornaments. 
 
 AMETHYST. Transparent. Of a rich violet or purple 
 colour. Used as a gem. 
 
250 DISCRIMINATION OF MINERALS. 
 
 EOSE QUARTZ. Seldom perfectly transparent. Of a 
 rosy tint. 
 
 CAIRNGORM, or SMOKY QUARTZ, is transparent, with a 
 smoky tinge. 
 
 FALSE TOPAZ has a yellow pellucid colour, distinguished 
 from topaz by its inferior hardness. 
 
 2. SEMI-TRANSPARENT VARIETIES : 
 
 CHALCEDONY. Pale colour and waxy lustre. Kesem- 
 bling icicles in some instances, the frothy surface of a 
 liquid in others. 
 
 CORNELIAN and SARD have red tints. 
 
 AGATE exhibits cloudy or moss-like patches, or a 
 number of lines arranged in circular and angular patterns. 
 
 ONYX or SARDONYX is made up of regular layers one 
 above another of different colours, often black, white, and 
 red. It is much used for cameos. 
 
 FLINT or HORNSTONE. Common dull varieties. 
 
 3. OPAQUE VARIETIES : 
 
 JASPER is quartz rendered opaque by clay, iron, and 
 other impurities ; it is of a red, yellow, or green colour ; 
 sometimes the colours are arranged in ribands, or in 
 other fantastic forms. It is used for ornamental work. 
 
 BLOODSTONE is green jasper with splashes of red re- 
 sembling blood-spots. 
 
 2. OPAL. 
 
 Chemical Composition. Silica and water. 
 
 Never crystallised. Fracture conchoidal. Specific 
 gravity, 2'2. Hardness, 6. Can be scratched by quartz, 
 and thus distinguished from it. 
 
 Infusible. It is generally milk-white. 
 
 Precious or Noble Opal exhibits a beautiful play of 
 colours, and is a valuable and rare gem. The common 
 varieties are of no value. 
 
 3. TALC. 
 
 Chemical Composition. Silica, magnesia, water (pro- 
 toxide of iron). 
 
DISCRIMINATION OF MINERALS. 251 
 
 Usually in irregular layers. Nearly opaque. White 
 or green ; pearly lustre ; greasy feel. Specific gravity, 
 2-7. Hardness, 1. Easily impressed by the nail. But 
 impure varieties are much harder. 
 
 Infusible. Yields no water when heated in a glass 
 tube. Is not attacked by boiling sulphuric acid. 
 
 Its greasy feel and pearly lustre readily distinguish it. 
 
 Mica, which is often confounded with it, is not so soft, 
 has not a greasy feel, and can be split into very thin trans- 
 parent layers. Steatite is a variety often applied to useful 
 purposes. 
 
 4. CHLORITE. 
 
 Chemical Composition. Silica, magnesia, alumina, pro- 
 toxide of iron, water. 
 
 Often forming rocks. Opaque. Green of various 
 shades. Lustre pearly or dull. Hardness, 1 to 2. Very 
 easily cut with a knife. 
 
 Infusible. In glass tube yields water. Boiling sul- 
 phuric acid extracts from it magnesia, alumina, and fer- 
 rous oxide. ' 
 
 Abundant. Of no value. 
 
 5. SERPENTINE. 
 
 Chemical Composition. Silica, magnesia, water (ferrous 
 oxide). 
 
 Often 'forming rocks. Never crystallised. Opaque. 
 Green. Lustre resinous or dull. Streak white. Hard- 
 ness, 3. Can be cut with a knife. Specific gravity, 2'5. 
 
 Infusible except in thin edges ; turns white in blowpipe 
 flame. Powder decomposed by sulphuric acid like chlorite. 
 Gives off water in glass tube. 
 
 Some varieties form handsome stone for slabs and 
 ornamental work. 
 
 Meerschaum, which is soft and earthy, and nephrite 
 (the New Zealand Maori greenstone), which is as hard as 
 quartz, both contain silica and magnesia. 
 
252 DISCRIMINATION OF MINERALS, 
 
 0. AUGITE AND HORNBLENDE. 
 
 Chemical Composition. Silica, lime, magnesia, ferrous 
 oxide (occasionally alumina and fluorine). 
 
 In four, six, or eight-sided prisms, exhibiting cleavage 
 in some directions. Usually opaque. Green, black, or 
 white. Glassy, pearly, or resinous lustre. Specific gravity, 
 3 to 3-5. Hardness, 5 to 6. Can be scratched with a 
 knife, using pressure. Scratched by quartz. 
 
 Dark varieties fusible. 
 
 Augite usually occurs in stout six or eight- sided prisms 
 with roof- like terminations ; hornblende in long slender 
 prisms. Asbestos is a variety composed of separable fibres 
 like flax. 
 
 Abundant in igneous and other rocks. Of no value. 
 Asbestos is used as a fireproof material, and the long silky 
 variety is woven into fireproof fabrics. 
 
 7. CHRYSOLITE, OR OLIVINE. 
 
 Chemical Composition. Silica, magnesia, protoxide of 
 iron. 
 
 In prisms ; but usually in grains or lumps resembling 
 bottle glass. Transparent or opaque. Green, yellow, or 
 black. Glassy lustre. Fracture conchoidal. Hardness, 
 6 to 7. Cannot be scratched with a knife. Specific 
 gravity, 3-4. 
 
 Infusible. 
 
 In appearance, hardness, and infusibility may resemble 
 tin ore, but is very much lighter in weight. Differs from 
 obsidian (volcanic glass) in its infusibility and superior 
 hardness. 
 
 Occurs in basalt and lava. Of no value. 
 
 8. TOURMALINE. . 
 
 Chemical Composition. Silica, alumina, magnesia, 
 boracic acid, fluorine, oxides of iron (lime and alkalies). 
 
 In prisms, with three, six, nine, or more sides, fur- 
 rowed lengthwise, terminating in low pyramids. Com- 
 
DISCRIMINATION OF MINERALS. 25$ 
 
 monly black and opaque ; rarely transparent, and of a rich 
 red, yellow, or green colour. Glassy lustre. Fracture 
 uneven. Specific gravity, 3vl. Hardness, 7 to 8. Can- 
 not be scratched with a knife. Not scratched by quartz. 
 
 Infusible. 
 
 When the smooth side of a prism is rubbed on cloth it 
 becomes electric and can attract a small piece of paper ; if 
 the prism is as wide as a pipe-stem, when one side is heated 
 for a moment in the blowpipe flame the opposite side be- 
 comes electric and can attract paper until the heat spreads 
 uniformly through the crystal. (On this account tour- 
 maline is said to be pyro- electric.) 
 
 Occurs in granite and slate. Of no value ; except the 
 fine-coloured transparent varieties, which are used as gems 
 and for optical purposes. 
 
 9. GARNET. 
 
 Chemical Composition. Silica, alumina, lime, iron, mag- 
 nesia, manganese. 
 
 Crystallised in dodecahedrons, never in prisms. Trans- 
 parent or opaque. Generally red ; also brown, green, 
 yellow, black, white. Glassy or resinous lustre. Fracture 
 conchoidal or uneven. Specific gravity, 3'5 to 4-3. Hard- 
 ness, 6-5 to 7'5. Cannot be scratched with a knife. 
 
 Fusible with more or less difficulty. Eed varieties 
 impart a green colour to borax bead owing to presence of 
 chromium. 
 
 Common in gneiss and schists. Fine-coloured trans- 
 parent varieties (carbuncle, cinnamon stone, almandine) 
 are used in jewellery. 
 
 10. TOPAZ. 
 
 Chemical Composition. Silica, alumina, fluorine. 
 
 In prisms, ''sometimes furrowed lengthwise, variously 
 terminated, breaking easily across with smooth brilliant 
 cleavage. 
 
 Transparent or semi-transparent. White, yellow, 
 greenish, bluish, pink. Glassy lustre. Specific gravity, 
 
254 DISCRIMINATION OF MINERALS. 
 
 3'5. Hardness, 8. Scratches quartz. Is scratched "by 
 sapphire. 
 
 Infusible, but often blistered and altered in colour by 
 heat. When smooth surfaces are rubbed on cloth they 
 become strongly electric and can attract small pieces of 
 paper, but rough surfaces do not show this. The brilliant 
 cleavage of topaz distinguishes it from tourmaline and 
 other minerals. 
 
 Occurs in granite. Used in jewellery. The topaz be- 
 comes electric by friction much easier than other gems, 
 such as the balas ruby, which it may resemble. The white 
 topaz resembles the diamond ; but, unlike diamond, it can 
 be scratched by sapphire. 
 
 11. BERYL, OR EMERALD. 
 
 Chemical Composition. Silica, alumina, glucina. 
 
 In six-sided prisms. Usually green. Transparent or 
 opaque. Glassy lustre. Fracture uneven. Specific gravity, 
 2'7. Hardness, 7 to 8. Scratches quartz. 
 
 Infusible, or nearly so, but becomes clouded by heating. 
 
 Occurs in granite and schist. Valuable for jewellery 
 when transparent and rich glass green (emerald), or sea 
 green (aquamarine). Opaque crystals of large size, exceed- 
 ing a ton in weight, have been found in North America. 
 
 12, ZIRCON, OR HYACINTH. 
 
 Chemical Composition. Silica, zirconia. 
 
 In square prisms, terminated by pyramids, and in octa- 
 hedrons. Often found in pebbles and grains. Transparent 
 or opaque. Wine or brownish red, grey, yellow, white. 
 Glassy lustre. Fracture usually irregular, but in one 
 direction it can be split so as to exhibit a smooth, even 
 cleavage face having an adamantine lustre like the diamond. 
 Specific gravity, 4'0 to 5'0. Hardness, 7'5. Scratches 
 quartz ; is scratched by topaz. 
 
 Infusible ; the red varieties, when heated before the 
 blowpipe, emit a bluish phosphorescent light, and become 
 permanently colourless. 
 
DISCRIMINATION OF MINERALS. 255 
 
 Occurs in syenite, granite, basalt. Clear crystals used 
 in jewellery, in jewelling watches, and in imitation of 
 diamond. It may be distinguished from diamond by its 
 inferior hardness, and in not becoming so readily electric 
 by friction. 
 
 13. FELSPAR. 
 
 Chemical Composition. Silica, alumina, potash or soda 
 (lime). 
 
 Crystallised or in irregular masses. Opaque. Usually 
 flesh-red, or white, or of various dull tints. Glassy or 
 pearly lustre. Fracture irregular ; but in some directions 
 it splits with an even, glimmering, cleavage face. Specific 
 gravity, 2-3 to 2-8. Hardness, 6. Easily scratched by- 
 quartz. Cannot be scratched with a knife without greatest 
 pressure. 
 
 In thin edges fusible with difficulty. 
 
 Abundant in granitic and porphyritic rocks. No value. 
 By its decomposition it forms porcelain clay or kaolin. 
 
 14. MICA. 
 
 Chemical Composition. Silica, alumina, magnesia, 
 potash, iron. 
 
 Always crystallised in thin plates, which may be split 
 into extremely thin flexible layers. Transparent in thin 
 layers. Brown or black. Lustre glassy, pearly, or 
 metallic. Streak white. Specific gravity, 2 -7 to 3*1. 
 Hardness, 2 to 2- 5. Very easily scratched with a knife. 
 
 Infusible. Differs from talc in not having a greasy 
 feel, in being harder, flexible, and affording thinner layers 
 perfectly transparent. 
 
 Abundant in granite and schist ; fine particles common 
 in sandstone. Applied to various uses when in large plates, 
 otherwise of no value. Was formerly used instead of glass 
 for windows. 
 
 15. ZEOLITES. 
 
 This name is used for a large class of silicates, com- 
 prising from fifty to a hundred different minerals, which 
 all contain water as an essential ingredient, and which 
 
256 DISCRIMINATION OF MINERALS. 
 
 melt readily before the blowpipe, and boil up owing to 
 the disengagement of steam. They occur filling pores 
 and cavities in basalt, lava, and other rocks. They are 
 usually white and well crystallised. Can be scratched with 
 a knife. Of no value. Silica, alumina, lime, soda, potash, 
 and water are the principal ingredients. 
 
 16. CORUNDUM, OR SAPPHIRE. 
 
 Chemical Composition. Alumina. 
 
 In six-sided prisms, often irregularly shaped. Some- 
 times in granular masses. Transparent or opaque. Blue, 
 black ; also red, green, yellow, white. Glassy lustre, some- 
 times pearly. Fracture uneven or conchoidal. Specific 
 gravity, 3-9 to 4'2. Hardness, 9. Easily scratches topaz 
 and quartz. In hardness it is only inferior to the diamond. 
 Infusible. 
 
 Occurs in river sands ; in granite, felspar, magnetic iron, 
 basalt. As a gem it stands next in value to the diamond r 
 but its tint must be brilliant and clear. The blue variety 
 is called Sapphire, the most esteemed shade being deep 
 velvet blue ; the blood-red variety is the Oriental ruby, 
 which can be easily distinguished from other red gems by 
 its superior hardness ; the bright yellow variety is the 
 Oriental topaz, distinguished by its hardness from the topaz, 
 yellow tourmaline, and false topaz ; the bright green is the 
 Oriental emerald ; the bright violet, Oriental amethyst ; 
 these varieties readily scratch the emerald (No. 11) and 
 amethyst (see quartz, No. 1) ; one variety exhibits a six- 
 rayed star inside the prism, and it is called the Asterias. 
 Dull crystals are called corundum, and grey or black 
 granular varieties emery ; these two kinds are used for 
 polishing powder. The ruby is the most highly prized 
 form of this mineral. 
 
 17. SPINEL. 
 
 Chemical Composition. Alumina, magnesia. 
 
 In octahedrons, occurring only crystallised. Usually 
 red and transparent ; also white, blue, green, yellow, brown, 
 black; the dark shades usually opaque. Lustre glassy. 
 
DISCRIMINATION OF MINERALS. 257 
 
 Fracture conchoidal. Specific gravity, 3*5 to 4*0. Hard- 
 ness, 8. Scratches quartz. 
 
 Infusible, and thus distinguished from garnet, which it 
 may resemble. Colour altered transiently by heat. Dis- 
 tinguished from zircon by its superior hardness and inferior 
 specific gravity. 
 
 Occurs in river sand ; in igneous rocks, gneiss, lime- 
 stone. The bright transparent varieties are used in jewel- 
 lery. When red it forms the common spinel, or balas- 
 ruby, which is distinguished from the Oriental ruby by 
 its inferior hardness ; bright green, chlorospinel ; orange, 
 rubicelte ; violet, almandine-ruby ; black, pleonast. 
 
 18. CHRYSOBERYL. 
 
 Chemical Composition. Alumina, glucina. 
 
 In prisms or tables. Transparent or semi-transparent. 
 Green. Lustre glassy. Fracture conchoidal ; imperfect 
 cleavage. Specific gravity, 3*5 to 3-8. Hardness, 8'5. 
 
 Infusible and unaltered before the blowpipe. 
 
 Distinguished from beryl by its specific gravity, its 
 tabular crystallisation, and its entire infusibility. 
 
 Occurs with beryl in river sand, gneiss, and granite. 
 Pellucid and fine opalescent varieties are used as gems. 
 
 19. DIAMOND. 
 
 Chemical Co mposition. Carbon . 
 
 In octahedrons, tetrahedrons, dodecahedrons, and forms 
 related to these ; the faces of the crystal sometimes curved. 
 Transparent. Colourless, yellow, red, green, blue, white, 
 brown, or black. Lustre adamantine. Breaks with smooth 
 cleavage planes parallel to the octahedral faces. Specific 
 gravity, 3*5 ; loses I0-35ths of its weight in water. Hard- 
 ness, 10. It is the hardest substance known, and scratches 
 all other minerals and gems. 
 
 Infusible. It burns and is consumed at a high tempe- 
 rature. 
 
 Becomes strongly electric when rubbed, and can then 
 attract light objects ; other gems do not exhibit this pro- 
 
 s 
 
258 DISCRIMINATION OF MINERALS. 
 
 perty unless polished. Some varieties, after exposure to 
 the sun, are said to give out light when placed in the 
 dark. 
 
 The distinguishing characters of diamond are, its hard- 
 ness, crystalline form, clean fracture, its brilliant reflection 
 and adamantine lustre, the facility with which it may be 
 electrified by friction, and a peculiar sound on rubbing 
 together. 
 
 Diamonds have been found in quartz- conglomerate and 
 micaceous sandstone, but are mostly obtained in river 
 beds. 
 
 20. GRAPHITE, OR BLACKLEAD. 
 
 Chemical Composition. Carbon . 
 
 In six-sided prisms ; but usually in uncrystallised wavy 
 layers. Opaque. Black. Lustre metallic. Specific gravity, 
 2. Hardness, 1 to 2. Very easily cut with a knife. Has 
 a greasy feel ; marks paper like a lead pencil. 
 
 Infusible. Burns slowly away. 
 
 Molybdenite and foliated tellurium resemble graphite ; 
 the former has a paler colour than graphite, and the latter 
 is very easily fusible. 
 
 Occurs in gneiss and slate. Valuable for lead pencils 
 and crucibles. 
 
 21. COAL. 
 
 Coal and carbonaceous shale differ from all the minerals 
 which they may resemble by burning before the blowpipe 
 and leaving a white or brown ash. The quantity of ash 
 affords an estimate of the purity of the coal. 
 
 22. APATITE. 
 
 Chemical Composition. Phosphoric acid, lime, fluorine. 
 
 In six-sided prisms. Also in masses. Transparent or 
 opaque. Usually green. Sometimes white, yellow, blue, 
 red, brown. Lustre resinous. Fracture conchoidal or 
 uneven. Specific gravity, 3*2. Hardness, 5. Can be 
 scratched with a knife, using pressure. 
 
 Infusible, except on very thin edges. Some kinds 
 
DISCRIMINATION OF MINERALS. 259 
 
 phosphoresce when heated. The pure mineral in powder 
 dissolves slowly in nitric acid without effervescence. The 
 crystals may resemble beryl, which, however, is too hard 
 to be scratched with a knife. 
 
 Occurs in gneiss, slate, limestone. Of value from its 
 use in the manufacture of artificial manures. 
 
 23. FLUOR-SPAR. 
 
 Chemical Composition. Fluorine, calcium. 
 
 In cubes or octahedrons. Also in masses. Transparent 
 or opaque. White or light violet, blue, green, or yellow; 
 sometimes layers of different tints in the same piece. Lustre 
 glassy. Breaks with smooth cleavage planes parallel to 
 the octahedral faces. Specific gravity, 3'0 to 3*2. Hard- 
 ness, 4. Can be scratched with a knife, but not so easily as 
 calcite. 
 
 Fusible with difficulty ; generally flies to pieces when 
 heated. Some varieties phosphoresce. 
 
 Occurs in veins with lead and silver ores. Used in 
 etching glass, and as a flux in smelting ; sometimes for 
 ornaments, but is too brittle. Abundant in many countries, 
 and of little value. 
 
 24. CALC-SPAR, OR CALCITE. 
 
 Chemical Composition. Carbonic acid, lime. 
 
 In rhombohedrons and other crystalline forms. Also 
 massive, earthy, or fibrous. Transparent or opaque. White 
 when pure ; often tinted. Lustre glassy, or dull. Breaks 
 with smooth cleavage planes parallel to the rhombohedral 
 faces. Specific gravity, 2-5 to 2*8. Hardness, 3. Easily 
 scratched with a knife, streak white. 
 
 Infusible before the blowpipe, but emits a strong light. 
 When burned, as in a kiln, it forms quicklime. Effervesces 
 when vinegar is poured upon it. It completely dissolves 
 in nitric acid with rapid effervescence. 
 
 Calc-spar is one of the most abundant minerals ; it 
 occurs in cavities and veins of all kinds of rock. The term 
 
 8 2 
 
260 DISCRIMINATION OF MINERALS. 
 
 calc-spar or calcite is restricted to the crystallised variety. 
 In an imperfectly crystallised and compact form it exists in 
 large rocky masses and beds ; all marbles and limestones 
 consist of it, mixed more or less with impurities. Chalk 
 and stalactites are nearly pure calcium carbonate. All 
 varieties of calcium carbonate may be easily distinguished 
 by being scratched with a knife, giving a white streak 
 whatever the colour of the mass may be, by effervescing 
 with an acid, and by being infusible. 
 
 25. MAGNESITE. 
 
 Chemical Composition. Carbonic acid, magnesia. 
 
 In rhombohedrons. Also globular, compact, earthy. 
 Transparent or opaque. White, yellow, brown. Lustre 
 glassy. Cleavage parallel to rhombohedral faces. Specific 
 gravity, 2-8 to 3*0. Hardness, 4 to 5. 
 
 Infusible. Dissolves in nitric acid, with very slow 
 effervescence. Does not yield quicklime when burnt. 
 
 Occurs with serpentine and limestone. Used in pre- 
 paring Epsom salts. 
 
 26. DOLOMITE. 
 
 Chemical Composition. Carbonic acid, lime, magnesia. 
 
 In rhombohedrons, faces often curved. Often granular 
 or massive. White or dull- tinted. Glassy or pearly. 
 Specific gravity, 2-8 to 2-9. Hardness, 3-5 to 4. 
 
 Infusible. Effervesces in nitric acid, and dissolves 
 more slowly than calc-spar. Yields quicklime when burnt 
 Occurs in extensive beds of various ages like limestone. 
 Used as a building stone, and in the manufacture of Epsom 
 salts. Difficult to distinguish from calcite without chemical 
 analysis. 
 
 27. ARAGONITE. 
 
 Chemical Composition. Same as calc-spar. 
 It differs from calc-spar in its crystalline form, which 
 is usually difficult to discern. It often occurs in fibrous 
 
DISCRIMINATION OF MINERALS. 261 
 
 clusters or in tangled branches. Specific gravity, 2'9. 
 Hardness, 3-5 to 4-0. 
 
 It has the general characters of calc-spar, but may be 
 distinguished from it by falling to powder in the blowpipe 
 flame, as well as by its superior hardness. 
 
 28. ROCK SALT. 
 
 Chemical Composition. Sodium chloride. 
 
 Has the characters of ordinary table salt, but is more 
 or less impure. Occurs in beds interstratified with sand- 
 stones and clays, which are usually of a red colour and 
 associated with gypsum. In the county of Cheshire, 
 where salt mines are worked, the surface indications are 
 brine springs supporting a vegetation like that near the 
 sea-coast ; also occasional sinking of the soil caused by the 
 removal of the subterranean bed of salt, by spring water 
 in some cases, and by mining operations in others. Small 
 and unimportant quantities of salt are often found encrust- 
 ing various rocks in dry weather. 
 
 29. SOLUBLE SULPHATES. 
 
 Aluminium sulphate alone or combined with potassium 
 (alum), and magnesium sulphate (Epsom salts), are often 
 found encrusting rocks. They are soluble in water, and 
 easily distinguished by their taste. 
 
 30. NITRE. 
 
 Chemical Composition. Potassium nitrate. 
 
 Nitre or saltpetre is another soluble mineral. It has a 
 cooling taste. It can be easily distinguished by the vivid 
 manner in which it burns on red-hot charcoal. 
 
 31. GYPSUM, SELENITE, OR ALABASTER. 
 
 Chemical Composition. Sulphuric acid, lime, water. 
 In prisms with oblique terminations ; sometimes resem- 
 bling an arrowhead. Transparent or opaque. White or 
 
262 DISCRIMINATION OF MINERALS. 
 
 dull-tinted. Glassy, pearly, or satin lustre. Cleavage 
 occurs easily in one direction. Specific gravity, 2-3. Hard- 
 ness, 2. Very easily cut with a knife. 
 
 Fusible with difficulty. In the blowpipe flame it be- 
 comes white and opaque without fusing, and can then be 
 easily crumbled between the fingers. Nitric acid does not 
 cause effervescence. 
 
 Occurs in fissures and in stratified rocks, often forming 
 extensive beds. When burnt it forms plaster-of-Paris ; it 
 is also used for ornaments, and as a manure. 
 
 32. HEAVY SPAK, OR BARYTES. 
 
 Chemical Composition. Sulphuric acid, baryta. 
 
 In tabular glassy crystals. Also in dull masses. Trans- 
 parent or opaque. White or tinted. Specific gravity, 4-3 
 to 4*8 ; its great comparative weight readily distinguishes 
 it. Hardness, 2-5 to 3 -5. 
 
 Splinters fly off the crystals when heated in the blow- 
 pipe flame. Fusible with difficulty. Not acted upon by 
 acids. 
 
 Occurs with various ores. Used as a white paint. 
 
 33. SULPHUR. 
 
 Crystallised or massive. Yellow. Eesinous lustre. 
 Specific gravity, 2-1. Hardness, T 5 to 2-5. Fusible; 
 burns with a blue flame and well-known odour. Occurs in 
 volcanic regions, and in beds of gypsum. 
 
 34. TIN ORE. 
 
 Chemical Composition. Tin, 78'4 ; oxygen, 21-6. 
 
 In four-faced prisms and pyramids, having an ada- 
 mantine lustre. Also in masses and grains (stream tin) 
 usually dull; sometimes resembling wood (wood tin). 
 Semi-transparent or opaque. Brown or black ; streak and 
 powder pale brown. Fracture uneven. Specific gravity, 
 6*8 to 7*0 ; the great comparative weight is an important 
 character to observe in distinguishing tin ore from other 
 
. DISCRIMINATION OF MINERALS. 263 
 
 minerals. Hardness, 6'0 to 7*0. Cannot be scratched 
 with a knife, and may thus be distinguished from blende, 
 which it resembles in lustre and infusibility. 
 
 Infusible. When mixed in powder with sodium carbon- 
 ate, placed on charcoal and covered with a small piece of 
 potassium cyanide, and then heated in the blowpipe flame, 
 a malleable globule of metallic tin is obtained. 
 
 Occurs in veins, and disseminated in granite, schist, 
 slate, and porphyry ; and in alluvial deposits. It is a 
 valuable ore, and the sole commercial source of the 
 metal. 
 
 35. MOLYBDENITE. 
 
 Chemical Composition. Molybdenum, 58*9 ; sulphur, 
 41-1. 
 
 In thin plates, like graphite (No. 20). Lustre metallic. 
 Colour, lead grey. Specific gravity, 4*5 to 4-6. Hardness, 
 1*0 to 1*5. Easily scratched by the nail. 
 
 Infusible. Tinges blowpipe flame faint green. Heated 
 on charcoal for a long time, it gives off a faint sulphurous 
 odour, and becomes encrusted white. 
 
 Occurs in granite, syenite, and chlorite schist. Not 
 applied to any particular uses. 
 
 36. BISMUTH. 
 
 Chemical Composition. Metallic bismuth. 
 
 Sometimes crystallised in rhombohedrons closely re- 
 sembling cubes, but generally massive. Lustre metallic. 
 White, with a tinge of red, liable to tarnish. Brittle. 
 Specific gravity, 9-6 to 9-8. Hardness, 2*0 to 2-5. Easily 
 scratched with a knife. 
 
 Easily fusible. Sometimes gives off an odour of garlic, 
 owing to admixture of arsenic. 
 
 Occurs with cobalt, silver, and tin ores in granite and 
 'slate rocks. Bismuth is a very valuable metal. 
 
 37. ANTIMONY SULPHIDE. 
 
 Chemical Composition. Antimony, 72*9 ; sulphur, 27*1. 
 Usually in long columnar or fibrous crystals. Also 
 
264 DISCRIMINATION OF MINERALS. 
 
 massive and granular. Lustre metallic. Lead colour. 
 Often tarnished. Streak metallic. Specific gravity, 4'6 to 
 4- 7. Hardness, 2. Very easily scratched with a knife. 
 
 Easily fusible. Before blowpipe gives off white vapours 
 and an odour of sulphur, and is entirely volatilised. When 
 the corner of a large piece of ore is fused, the border of 
 the fused part is often tinted red. When heated on char- 
 coal with cyanide of potassium, it gives a globule of metallic 
 antimony, which is brittle, has a crystalline surface, burns 
 when strongly heated, emitting white fumes, and can be 
 entirely volatilised. 
 
 Occurs in veins in granite and slate, alone, or with 
 ores of silver, lead, and other metals. This ore is the 
 principal commercial source of the metal. 
 
 38. ARSENIC. 
 
 Chemical Composition. Metallic arsenic. 
 
 Seldom distinctly crystallised. Usually in fine granular 
 or spherical masses. Colour white, usually with a black 
 tarnish. Streak white, metallic. Brittle. Specific gravity, 
 5-7 to 5-8. Hardness, 3-5. 
 
 Before the blowpipe it quickly volatilises without 
 fusing, giving off white fumes having an odour of garlic. 
 
 Occurs in veins with lead and silver ores. 
 
 39. ARSENIC SULPHIDE. 
 
 Chemical Composition. Sulphur, arsenic. 
 
 In crystals, or massive. Yellow or red. Semi-transparent 
 or opaque. Eesinous or glassy lustre. Specific gravity, 3*5. 
 Hardness, 1*5. Volatilised before the blowpipe with a 
 blue flame. 
 
 Occurs in veins with arsenical ores ; in beds of clay, 
 limestone, and gypsum, and has been observed in lava. 
 Used as a pigment and in making fireworks, but objection- 
 able for each of these uses, owing to its highly poisonous 
 character. 
 
DISCRIMINATION OF MINERALS. 2(35 
 
 40. NATIVE IRON. 
 
 Chemical Composition. Iron, with a small percentage 
 of nickel. 
 
 Occurs in meteorites. Resembles ordinary iron. Mal- 
 leable. Is attracted by a magnet. Specific gravity, 7*0 
 
 to 7-8. 
 
 41. IRON PYRITES. 
 
 Chemical Composition. Iron, 46 '7 ; sulphur, 53'3. 
 
 In cubes and allied forms ; sides often marked by fine 
 parallel lines. Also massive. Brass yellow. Lustre metallic. 
 Fracture irregular. Specific gravity, 4*8 to 5*1. Hardness, 
 6 to 6- 5. Cannot be scratched with a knife ; scratched by 
 quartz ; scratches glass with great facility. Strikes fire 
 with steel (the origin of the term pyrites). 
 
 Before the blowpipe it burns with a blue flame, giving 
 off an odour of sulphur, and ultimately fuses into a, black 
 magnetic globule. 
 
 Abundant. Used as a source of sulphur and sulphuric 
 acid ; occasionally auriferous. This ore and arsenical 
 pyrites form the ' mundic ' of miners. It is easily dis- 
 tinguished from copper pyrites by its hardness ; copper 
 pyrites being easily cut with a knife. Distinguished from 
 gold by its hardness and in not being malleable, and in 
 giving off* sulphurous odours in the blowpipe flame. 
 
 42. ARSENICAL PYRITES (MISPICKEL). 
 
 Chemical Composition. Iron, 34'4 ; arsenic, 19-6 ; 
 sulphur, 46 -0. 
 
 In flattened prisms. Also massive. White. Lustre 
 metallic. Streak grey. Fracture uneven. Specific gravity, 
 6-0 to 6*3. Hardness, 5- 5. Cannot be scratched with a 
 knife ; scratched by quartz. 
 
 Heated before the blowpipe it gives off white arsenical 
 fumes of a garlic odour, and ultimately fuses into a black 
 globule. 
 
 Abundant in mining districts ; sometimes auriferous. 
 
 This ore and iron pyrites form the ' mundic ' of miners. 
 
DISCRIMINATION OF MINERALS. 
 
 43. MAGNETIC IRON. 
 
 Chemical Composition. Iron, 72-4 ; oxygen, 27*6. 
 
 In octahedrons and dodecahedrons. Also in masses 
 (lodestone) and in grains. Black. Lustre metallic. Streak 
 or powder black. Fracture irregular. Specific gravity, 
 5-0 to 5-2. Hardness, 5-5 to 6 -5. Not scratched with a 
 knife. Magnetic ; it can attract iron filings. Is itself 
 attracted by a magnet. 
 
 Infusible. With borax bead gives the indications of 
 iron. 
 
 Occurs in many rocks, sometimes in beds, or forming 
 mountainous masses ; common in river sands. Used as an 
 ore of iron. 
 
 44. SPECULAR IRON, HEMATITE, OR MICACEOUS IRON. 
 
 Chemical Composition. Iron, 70 ; oxygen, 30. 
 
 In tabular crystals or scales. Also fibrous, massive, 
 granular, earthy. Colour black. Streak or powder, red. 
 Lustre, metallic or dull. Specific gravity, 4*5 to 5-3. 
 Hardness of crystals, 5-5 to 6*5. Not scratched with a 
 knife. Earthy varieties softer, and can be scratched with 
 a knife. 
 
 Infusible. With borax bead gives the indications of 
 iron. An abundant ore of iron. Oft en gradually changes 
 into red or brown ochre. 
 
 45. RED FERRIC OXIDE, OR RED OCHRE. 
 
 Chemical Composition. Iron oxide, and more or less 
 water. 
 
 An uncrystalline earthy variety of the preceding, often 
 mixed with clay. Colour, bright or dull red. Can generally 
 be scratched with a knife. 
 
 Blackens when heated, but regains its red colour on 
 cooling. With borax bead gives the indications of iron. 
 Abundant ore of iron. 
 
 
DISCRIMINATION OF MINERALS. 267 
 
 4G. BROWN FERRIC OXIDE (YELLOW OR BROWN OCHRE). 
 
 Chemical Composition. Iron oxide, water. 
 
 Like the last, but of a brown, yellow, or black colour. 
 Eartliy, fibrous, stalactitic. Scratched with a knife. 
 
 Blackens before the blowpipe. With borax bead gives 
 the indications of iron. An abundant ore of iron. 
 
 47, TITANIC IRON. 
 
 Chemical Composition. Iron oxides and titanic acid in 
 variable proportions. 
 
 In octahedrons or in tabular plates. Also in grains. 
 Black. Lustre metallic. Streak or powder black. Specific 
 gravity, 4- 5 to 5*3. Hardness, 5 to 6-5. Not scratched 
 with a knife. 
 
 Infusible. With borax gives the indications of iron. 
 With microcosmic salt, which is often used instead of borax 
 in an exactly similar way, it gives a red bead in the 
 reducing part of the flame, but rather a large quantity 
 of the mineral must be used to obtain this result. 
 
 It is sometimes magnetic. 
 
 Its black streak or powder distinguishes it from 
 specular iron, which it often resembles. Common in some 
 river sand. 
 
 48. CHROMIC IRON. 
 
 Chemical Composition. Chromium sesquipxide, ferrous 
 oxide (alumina, magnesia). 
 
 In octahedrons. Usually massive. Black. Lustre 
 faintly metallic. Streak or powder dark brown. Fracture 
 irregular. Specific gravity, 4-4 to 4'6. Hardness, 5'5. 
 Not scratched with a knife. 
 
 Infusible. With borax bead gives the characteristic 
 indications of chromium. 
 
 Occurs in serpentine. Used in the preparation of 
 chromium colours. * 
 
 49. GREENEARTH. 
 
 Chemical Composition. Iron silicate, and other ingre- 
 dients. Has a green earthy appearance, often resembling 
 
268 DISCRIMINATION OF MINERALS. 
 
 an ore of copper, but is readily distinguished from copper 
 by blowpipe tests, and by not forming a blue solution in 
 nitric acid. 
 
 50. IRON CARBONATE. 
 
 Chemical Composition. Carbonic acid, 37 '9; ferrous 
 oxide, 62-1. 
 
 In rhombohedrons ; faces often curved. Usually 
 massive, globular, fibrous, or encrusting. Light or dark 
 brown. Glassy or pearly lustre. Streak white. Specific 
 gravity, 3 -7 to 3 -9. Hardness, 3*5 to 4-5. Scratched with 
 a knife. 
 
 Infusible. Blackens when heated. With borax bead 
 gives the indications of iron. Dissolves in nitric acid with 
 effervescence when heated. 
 
 Occurs in beds and nodules in stratified rocks ; in veins 
 and cavities. It is often mixed with clay (clay ironstone). 
 Abundant ore of iron. 
 
 61. MANGANESE ORES. 
 
 Chemical Composition. Various manganese oxides. 
 
 Crystallised or massive. Black. Lustre unmetallic ; 
 dull or shining. Powder or streak brown or black. Spe- 
 cific gravity, 4 to 5. Hardness generally below 3. Very 
 easily scratched with a knife. 
 
 Infusible. With borax bead gives the characteristic 
 indications of manganese. Widely distributed. Used in 
 chemical manufactures. 
 
 52. ARSENICAL NICKEL. 
 
 Chemical Composition. Nickel, 44 ; arsenic, 56. 
 
 Usually in masses of a pale copper colour and metallic 
 lustre. Specific gravity, 7*2 to 7*8. Hardness, 5 to 5-5. 
 Scratched with a knife, using pressure. 
 
 Before the blowpipe on charcoal it melts, giving out 
 white arsenical fumes having a garlic odour. It is readily 
 distinguished by its pale copper red colour and its arsenical 
 fumes when heated. 
 
DISCRIMINATION OF MINERALS. 26$ 
 
 Occurs in veins in granite and slate, with ores of cobalt, 
 silver, copper, bismuth, lead. A valuable source of metallic 
 nickel. 
 
 53. SMALTINE (TIN-WHITE COBALT). 
 
 Chemical Composition. Cobalt up to 24 per cent., 
 arsenic. 
 
 In octahedrons, cubes, dodecahedrons, and allied forms. 
 Also massive. Tin-white or steel-grey. Lustre metallic. 
 Streak greyish -black. Fracture uneven. Specific gravity, 
 6-3 to 6-6. Hardness, 5-5 
 
 Fusible. In the blowpipe flame gives off arsenical 
 fumes (odour of garlic). With borax bead gives the 
 characteristic indications of cobalt. In nitric acid forms 
 a pink solution. Eesembles mispickel and iron pyrites, 
 but is at once distinguished by the test with borax bead. 
 Its arsenical fumes distinguish it from iron pyrites, and its 
 crystalline form from mispickel. 
 
 Occurs in veins in slate and gneiss. A valuable ore of 
 cobalt. 
 
 54. COBALT BLOOM. 
 
 Chemical Composition. Cobalt oxide, 37*6 ; arsenic 
 acid, 384 ; water, 24'0, 
 
 In oblique crystals, with a highly perfect cleavage like 
 mica. Also in incrustations. Eed or pink, grey, green. 
 Lustre brilliant pearly. Transparent or opaque. Specific 
 gravity, 2-9 to 3-1, Hardness, 1-5 to 2. Very easily cut 
 with a knife. 
 
 Fusible in blowpipe flame, evolving arsenical fumes. 
 When heated on charcoal it gives off an odour of arsenic. 
 With borax bead gives indication of cobalt. 
 
 Occurs in beds and veins with other ores of cobalt. A 
 valuable ore of cobalt. 
 
 55. BLENDE. 
 
 Chemical Composition. Zinc, 66*7 ; sulphur, 33 -3. 
 In dodecahedrons, octahedrons, and allied forms. Also 
 massive. Yellow, red, brown, black. Lustre adamantine, 
 
270 DISCRIMINATION OP MINERALS. 
 
 resinous, or waxy. Transparent or opaque. Breaks with 
 brilliant cleavage faces in some directions. Specific 
 gravity, 4*0 to 4*1. Hardness, 3 -5 to 4*0. Easily scratched 
 with a knife. 
 
 Infusible. Emits a strong light when heated, but no 
 odour of sulphur is perceptible. It is easily distinguished 
 by its waxy lustre, softness, infusibility, and perfect cleav- 
 age. It dissolves at once in nitric acid. 
 
 Occurs with lead and copper ores. It is the 'black 
 jack' of miners. An ore of zinc, but more difficult to 
 smelt than the carbonate and silicate. 
 
 56. ZINC CARBONATE (CALAMINE). 
 
 Contains 52 per cent, of zinc. 
 
 Usually in crusts or masses. White, green, or brown. 
 Opaque. Pearly or glassy. Specific gravity, 4*1 to 4'5. 
 Hardness, 5. Can be scratched with a knife, using a little 
 pressure. 
 
 Infusible. On charcoal becomes yellow whilst hot, 
 white on cooling. Dissolves rapidly with effervescence 
 when- heated with nitric acid. 
 
 Occurs with galena and blende. A valuable zinc ore. 
 
 57. ZINC SILICATE (SMITHSONITE). 
 
 Contains 53 per cent, of zinc. 
 
 In prisms, or massive. White, greenish, bluish, or 
 brownish. Glassy lustre. Transparent or opaque. Specific 
 gravity, 3- 3 to 3*5. Hardness, 5. 
 
 Infusible. Shines with a green light in the blowpipe 
 flame. Does not effervesce with nitric acid, but dissolves, 
 leaving a jelly of silica. 
 
 Occurs with zinc carbonate. A valuable zinc ore. 
 
 58. GALENA. 
 
 Chemical Composition. Lead, 86*6 ; sulphur, 13*4. 
 
 In cubes. Also granular, massive. Lead colour. 
 Metallic lustre. Streak metallic. Breaks into cubical 
 fragments with bright cleavage faces. Specific gravity, 
 
DISCRIMINATION OF MINERALS. 271 
 
 7*4 to 7'7. Hardness, 2-5. Very easily scratched with a 
 knife. 
 
 Easily fusible. Before the blowpipe on charcoal is 
 reduced to a metallic globule of lead, giving off an odour 
 of burning sulphur. 
 
 Occurs in granite and stratified rocks. Often associated 
 with copper and other ores. The principal ore of lead. 
 It usually contains a small quantity of silver. 
 
 59. LEAD CAKBONATE (CERUSITE). 
 
 Contains 77 per cent, of lead. 
 
 In prisms, sometimes united in four or six-rayed crosses. 
 White or grey'." Transparent or opaque. Lustre glassy. 
 Specific gravity, 6-4 to 6 -6. Hardness, 3- 5. 
 
 Flies violently to pieces in the blowpipe flame. If 
 placed in a cavity on charcoal, and covered with sodium 
 carbonate, then carefully fused by the flame, it yields a 
 globule of metallic lead. In nitric acid it dissolves with 
 effervescence. 
 
 Usually occurs with galena. It is a valuable lead ore. 
 
 60. LEAD SULPHATE (AJSTGLESITE). 
 
 Contains 68 per cent, of lead. 
 
 In slender brilliant crystals upon galena. Also massive. 
 White or grey. Transparent or opaque. Specific gravity, 
 6-3. Hardness, 3. 
 
 Before the blowpipe fusible, but apt to decrepitate on 
 charcoal ; with sodium carbonate yields a globule of me- 
 tallic lead. Differs from lead carbonate in not dissolving 
 with effervescence in nitric acid. 
 
 Usually occurs with galena, and results from its decom- 
 position. 
 
 61. PYROMORPHITE (LEAD PHOSPHATE). 
 
 Chemical Composition. Lead oxide, 74'0 ; phosphoric 
 acid, 15-8 ; lead chloride 10-2. 
 
 In stout prisms, grouped together. Also massive. 
 Bright green or brown. Opaque or semi transparent. 
 
272 DISCRIMINATION OF MINERALS. 
 
 Lustre resinous. Streak white. Fracture irregular. 
 Specific gravity, 6*9 to 7*1. Hardness, 3-5 to 4-0. Easily 
 scratched with a knife. 
 
 Easily fusible. With sodium carbonate on charcoal 
 the lead is reduced. Soluble in nitric acid. 
 
 Occurs sparingly in veins with galena. 
 
 62. CINNABAR. 
 
 Chemical Composition. Mercury, 86'2 ; sulphur, 13 -8. 
 
 In granular, compact, and earthy masses. Sometimes 
 in crystals, exhibiting adamantine cleavage faces. Opaque 
 or semi-transparent. Vermilion or brownish red. Specific 
 gravity, 8-0 to 8-2. Hardness, 2-5. Very easily scratched 
 with a knife. 
 
 Before the blowpipe it volatilises, giving off a strong 
 odour of burning sulphur. Mixed with dried sodium car- 
 bonate, and heated over a candle-flame, in an iron spoon, 
 it gives off vapours of mercury, which may be condensed 
 on a gold coin held half an inch above the mixture. The 
 surface of the coin appears whitish at first, but when 
 rubbed between the fingers becomes brilliantly amalga- 
 mated. With care, this test easily detects one per cent, 
 of cinnabar in an ore. The mercury is removed from the 
 gold coin by gentle heating. The blowpipe tests distin- 
 guish it at once from red oxide of iron and all other red 
 minerals. 
 
 Occurs in talcose and argillaceous rocks. It is the 
 principal source of the mercury of commerce. 
 
 63. NATIVE MERCURY, OR QUICKSILVER. 
 
 The metal in a pure state is rarely found. It occurs 
 disseminated in liquid globules through sandstone and 
 other rocks, in cavities of which it may accumulate. It 
 is easily recognised. A rock suspected to contain mer- 
 cury may be tested by simply heating it as described 
 under cinnabar ; but without the addition of carbonate 
 of soda. 
 
DISCRIMINATION OF MINERALS. 273 
 
 64. NATIVE COPPER. 
 
 Usually in strings, plates, or irregular masses : some- 
 times crystalline. Like ordinary copper, but often tar- 
 nished. Specific gravity, 8-9. Easily scratched with a 
 knife. Malleable, can be flattened out under a hammer. 
 
 Occurs with copper ores. 
 
 65. VITREOUS COPPER. 
 
 Chemical Composition. Copper, 79*8 ; sulphur, 20*2. 
 
 Sometimes in prisms, but usually massive. Blackish lead 
 grey, tarnished. Streak metallic. Specific gravity, 5'5 to 
 5 -8. Hardness, 2-5 to 3. Very easily scratched with a knife. 
 
 Fusible. Before the blowpipe gives off an odour of 
 sulphur. When heated on charcoal, a malleable globule 
 of metallic copper remains, tarnished black, but rendered 
 evident on flattening under a hammer. With borax bead 
 gives the indications of copper. Dissolves in nitric acid, 
 forming a blue solution. (These tests distinguish it from 
 sulphide of silver.) 
 
 Occurs with other copper ores. A valuable ore of copper. 
 
 66. COPPER PYRITES. 
 
 Chemical Composition. Copper, 34*6 ; iron, 30'9 ; sul- 
 phur, 34-9. 
 
 In tetrahedrons or octahedrons. Usually massive. 
 Brass yellow, often tarnished. Lustre metallic. Streak 
 unmetallic, blackish green. Fracture uneven. Specific 
 gravity, 4*1 to 4*3. Hardness, 3'5 to 4'0. Easily scratched 
 with a knife. 
 
 Fusible. Gives off an odour of sulphur before blow- 
 pipe. Does not give the indications of copper with borax 
 bead, or when heated upon charcoal with sodium carbon- 
 ate. Dissolves in nitric acid, forming a blue solution. 
 Distinguished from iron pyrites by being easily cut with a 
 knife ; and from gold by not flattening under a hammer, 
 and by its greenish powdery streak. 
 
 Occurs in granite and slate in lodes or veins. Valuable 
 ore of copper. 
 
 T 
 
274 DISCRIMINATION OF MINERALS. 
 
 07. GREY COPPER. 
 
 This term includes a variety of ores having a common 
 crystalline form, generally the tetrahedron also ; a definite 
 chemical formula, though the ingredients are numerous 
 and may be variously combined within certain limits. 
 Sulphur is an invariable ingredient ; and arsenic or anti- 
 mony, one or both, must be present ; the other ingredients 
 are copper, iron, zinc, lead, silver, or mercury, in variable 
 proportions. The copper ranges up to 40 per cent. ; and 
 in some kinds as much as 30 per cent, of silver has been 
 found. It also occurs massive. Steel grey to iron black. 
 Lustre metallic. Streak black, or dark red when zinc is 
 present. Fracture uneven. Specific gravity, 4'5 to 5-2. 
 Hardness, 3 to 4. Can be easily scratched with a knife. 
 
 Fusible. Before the blowpipe gives off an odour of 
 sulphur, also white inodorous fumes of antimony, and 
 occasionally arsenic. 
 
 Copper cannot be detected by the blowpipe tests. It 
 dissolves in nitric acid, forming a greenish brown solution. 
 
 Occurs with copper pyrites, galena, and blende. This 
 ore is wrought for copper, and occasionally for silver. 
 
 68. BLACK CUPRIC OXIDE. 
 
 Heavy black powder or mass. Soft. Easily distin- 
 guished from manganese by affording the indications of 
 copper by the blowpipe tests. It results from the waste 
 of various copper ores. Valuable as a source of the 
 metal. 
 
 69. RED CUPROUS OXIDE. 
 
 Chemical Composition. Copper, 88-8 ; oxygen, 11*2. 
 
 In octahedrons and dodecahedrons. Also in granular 
 and earthy masses. Eed. Lustre adamantine, metallic, 
 or earthy. Streak red. Semi-transparent or opaque. 
 Exhibits cleavage parallel with octahedral faces. Specific 
 gravity, 6. Hardness, 3'5 to 4-0. Can be scratched 
 with a knife. 
 
 Before the blowpipe on charcoal it yields a globule of 
 metallic copper. With borax bead gives the indications 
 
DISCRIMINATION OF MINERALS. 275 
 
 of copper. Forms a blue solution in nitric acid. These 
 tests distinguish it from red ferric oxide. 
 
 Occurs in granite and slate, with copper ores and 
 galena. Valuable source of the metal. 
 
 70. COPPER CARBONATES, BLUE AND GREEN. 
 
 Chemical Composition. Oxide of copper, carbonic acid, 
 water ; the percentage of metallic copper about 56. 
 
 In crystals, but usually in fibrous, silky, globular, 
 incrusting masses. Blue or green. Opaque. Glassy, silky, 
 or dull. Specific gravity, 3-7 to 4'0. Hardness, 3'5 to 
 4*0 ; can be scratched with a knife. 
 
 Blacken when heated. On charcoal are reduced to a 
 globule of pure copper. Give the indications of copper 
 with borax bead. Soluble in nitric acid with effervescence, 
 forming a blue solution. 
 
 Copper Silicate resembles the carbonate, and is distin- 
 guished by dissolving in nitric acid without effervescecce. 
 
 Occur with copper ores, and result from their decom- 
 position. Valuable sources of the metal. 
 
 71. PLATINUM. 
 
 In flattened or angular grains or nuggets, which are 
 malleable. Steel-grey. Lustre metallic. Specific gravity, 
 IT to 19. As heavy as gold, and, therefore, easily distin- 
 guished and separated from lighter materials. 
 
 Infusible. Insoluble in nitric acid. 
 
 Occurs in quartz veins, but principally in alluvial 
 deposits with gold. Used chiefly for chemical apparatus. 
 Of great value. 
 
 72. GOLD. 
 
 In dust, grains, or nuggets in river sand ; or in wiry, 
 branching, and irregular forms in quartz. Pale or deep 
 yellow. Malleable. Specific gravity, 15 to 19. Hard- 
 ness, 2-5 to 3-0. 
 
 Fusible without blackening, and without giving off 
 any odour. Imparts no colour to boiling nitric acid. 
 
 T 2 
 
270 DISCRIMINATION OF MINERALS. 
 
 The minerals commonly accompanying gold are iron 
 pyrites, arsenical iron, iron and manganese oxides, galena., 
 zinc blende, and copper pyrites in quartz veins ; and 
 magnetic iron, titanic iron, chromic iron, tin ore, quartz, 
 zircon, topaz, corundum, diamond, in alluvial deposits. 
 
 73. SILVER. 
 
 In strings, plates, and branching forms penetrating 
 quartz, porphyry, slate, granite. Silver white, but usually 
 tarnished black. Malleable. Specific gravity, about 1O5, 
 Hardness, 2-5 to 3v 
 
 Fusible without giving off any odour. Soluble in 
 nitric acid, and on adding salt to the solution a white 
 curd is thrown down which blackens on exposure to sun- 
 light. 
 
 74. SULPHIDE OP SILVER. 
 
 Chemical Composition. Silver, 87 ; sulphur, 13. 
 
 In dodecahedrons or allied forms. Also massive. 
 Black. Opaque. Lustre metallic. Streak shining. 
 Specific gravity,* 7*2. Hardness, 2-0 to 2'5. Very easily 
 cut with a knife. 
 
 Very fusible, giving off an odour of sulphur when 
 heated. Before the blowpipe on charcoal, with or with- 
 out sodium carbonate, it yields a white globule of metallic 
 silver which can be flattened under a hammer. 
 
 The ore .is soluble in nitric acid, and on adding salt to 
 the solution a white curdy precipitate is thrown down y 
 which blackens on exposure to sunlight. 
 
 Occurs in veins in granite, porphyry, and slate, with 
 arsenic, silver, and lead ores. 
 
 75. ANTIMONIAL AND ARSENICAL SILVER ORES. 
 
 Several ores of silver contain arsenic and antimony, 
 as well as sulphur ; the percentage of silver in these ores 
 varies from 12 to 68. Bed, grey, or black. Lustre ada- 
 mantine or metallic. Red streak. Specific gravity, 5 to 
 0. Hardness, 2 to 3. Easily scratched with a knife. 
 
DISCRIMINATION OF MINERALS. i>77 
 
 Fusible. Before the blowpipe give off an odour of 
 sulphur, or arsenical fumes of a garlic odour, or fumes of 
 antimony. Heated on charcoal with sodium carbonate, 
 afford a globule of metallic silver. 
 
 Nitric acid extracts the silver from these ores, forming 
 a solution in which salt throws down a white curd, black- 
 ening on exposure to sunlight. 
 
 76. HORN SILVER. 
 
 Chemical Composition. Silver, 75-3 ; chlorine, 24*7. 
 In veins with silver ores. Greenish. Waxy lustre. Hard- 
 ness, I'O to 1'5. Cuts like wax or horn. 
 
 Very easily fusible. Heated with sodium carbonate, 
 'On charcoal, it yields a globule of metallic silver. 
 
 77. LEAD AND ANTIMONY SULPHIDES. 
 
 Lead- or steel-grey. Lustre metallic. Hardness not 
 exceeding 3. Specific gravity about 6. 
 
 Fusible. Give off an odour of sulphur and white 
 fumes of antimony before the blowpipe ; and with sodium 
 carbonate on charcoal afford a bead of metallic lead. 
 
 78. MERCURY, LEAD, SILVER, OR COPPER SELEXIDE. 
 
 Gives off a strong odour of horse-radish, due to sele- 
 nium, when heated before the blowpipe. The metal may 
 be found in some cases by sodium carbonate on charcoal, 
 in others by the use of nitric acid. 
 
 79. MILLERITE (NICKEL SULPHIDE). 
 
 Chemical Composition. Nickel, 64- 9 ; sulphur, 35*1. 
 In delicate needles. Brass-yellow. Metallic lustre. 
 Fusible. Hardness, 3 to 5. Imparts to borax the 
 -colour of nickel. 
 
278 DISCRIMINATION OF MINERALS. 
 
 80. WHITE NICKEL. 
 
 Chemical Composition. Nickel, 28-3 ; arsenic, 71*7. 
 
 Same colour, hardness, specific gravity, and crystalline 
 form as smaltine, No. 53. Often contains cobalt, and 
 graduates into smaltine, with which it is usually found. 
 
 81. RUTILE. 
 
 Chemical Composition. Titanium and oxygen. 
 
 In crystals and masses. Eed brown. Streak paler. 
 Lustre sub-metallic. Hardness, 6'0 to 6-5. Specific 
 gravity, 4-2 to 4-3. 
 
 Infusible. With borax bead, yellowish-green or colour- 
 less (oxidising), dirty violet (reducing). 
 
 8-2. SPHENE. 
 
 Chemical Composition. Titanic acid, silica, lime. 
 
 In thin wedge-shaped crystals. Yellow, green, brown. 
 Transparent or opaque. Lustre adamantine. Hardness. 
 5 to 5*5. Specific gravity, 3-2 to 3*6. Transparent yellow, 
 bead with borax when hot. 
 
 Fusible with difficulty on edges. 
 
 83. WOLFRAM. 
 
 Chemical Composition. Tungstic acid, iron, manga- 
 nese. 
 
 Crystals or masses. Brownish black. Lustre shining 
 or dull. Opaque. Hardness, 55. Specific gravity, 7 
 to 8. 
 
 Fusible with difficulty. With borax gives the colour 
 of iron. Characterised by its great weight. Found often 
 with tin ores. 
 
 84. PITCHBLENDE. 
 
 'Chemical Composition. Uranium, 84'S ; oxygen, 15*2.. 
 
 Massive. Black, opaque. Resinous lustre. Streak 
 black. Hardness, 5-5. Specific gravity, 6*4 to 6*7. 
 
 Infusible. With borax gives yellow bead. Dissolves 
 in hot nitric acid, forming a yellow solution. 
 
DISCRIMINATION OF MINERALS. 279 
 
 85. MAGNETIC IRON PYRITES. 
 
 Chemical Composition. Iron, 60'5 ; sulphur, 39-5. 
 Usually massive. Bronze yellow, tarnished. Metallic 
 lustre. Feebly magnetic. Hardness, 3*5 to 4-5. 
 Fusible. 
 
 DETERMINATION OF MINERALS. 
 
 The following scheme* will show how a mineral may 
 be determined by means of the tests and characters 
 already indicated. 
 
 The scheme first separates minerals into two divisions 
 namely, those which possess a metallic lustre, and those 
 which do not ; and these are further subdivided into 
 groups by means of other characters. 
 
 The various members of each group are enumerated 
 in the tables, which also give the mode of discrimination. 
 
 In most cases it is easy to decide whether a mineral 
 possesses a lustre like a polished metal, but where the 
 lustre is less decided there is room for a difference of 
 opinion ; besides, a mineral may possess a metallic lustre 
 in some specimens, and not in others. 
 
 To provide against this, any mineral about which 
 difficulty may arise is placed in both divisions, and even in 
 several groups if necessary. The mineral blende affords 
 an instance. 
 
 From inspection of the scheme it will be observed that 
 valuable metalliferous minerals possess, as a rule, a metallic 
 lustre, are heavy in proportion ta their bulk, and are easily 
 scratched with a knife. The chief exception is tin ore, 
 which has not a metallic lustre, and which cannot be 
 impressed with a knife. 
 
 In testing it is necessary to bear in mind that a piece 
 of ore may consist of a single mineral only, either pure, or 
 mixed with a readily distinguishable gangue ; or it may be 
 composed of finely disseminated particles of two or more 
 metalliferous minerals. But by search in localities where 
 such mixtures occur, pure and isolated specimens of each 
 
 * Drawn up by Dr. A. M. Thompson. 
 
280 DETERMINATION OF MINERALS 
 
 kind may generally be found. Moreover, after a little 
 practice the complication of characters which mixtures 
 present will not interfere with the recognition of their 
 components. 
 
 The scheme is not expected to prove an infallible means 
 of determining minerals in all possible instances, though 
 previous experience has shown that it is of use to those 
 who have learned to apply the tests recommended in the 
 former part of this chapter. To avoid error, it is necessary 
 in all cases, before coming to a decision, for the observer 
 to be convinced that the mineral under examination ac- 
 tually possesses all the characters of the mineral which it 
 is supposed to be. 
 
 Serious mistakes may arise by drawing inferences from 
 the scheme only, and neglecting to confirm all the other 
 characters. In performing the experiments recommended, 
 the greatest care and accuracy are needful ; to the careless 
 and superficial observer, or whoever is not proof against 
 self-deception, the scheme can be of little use. 
 
 The characters of each mineral have been described in 
 the preceding pages. 
 
 SCHEME FOR THE DETERMINATION OF MINERALS. 
 
 DIVISION I. Minerals possessing a METALLIC lustre. 
 
 Experiment. Heat a fragment of the mineral supported 
 on charcoal, or held with a forceps, in the oxidising point of 
 the blowpipe flame. 
 
 1. An odour of burning sulphur is given off. (See 
 Group I.) 
 
 2. An odour of garlic is given off, or white fumes not 
 having an odour of sulphur. (See Group II.) 
 
 3. No fumes or odour given off. (See Group III.) 
 
 Note. An odour of horse-radish indicates Selenium (78). 
 
 DIVISION II. Minerals which do NOT poss'ess a metallic 
 lustre. 
 
 Experiment. Scratch the mineral with a knife or sharp 
 
BEFORE THE BLOWPIPE. 281 
 
 fragment of quartz or sapphire, according to its hardness, 
 and observe the colour of the streak. 
 
 Note. If the streak is metallic the unmetallic lustre is due to tarnish. See 
 Division I. 
 
 A. The streak is black or coloured. 
 
 Experiment. Hold a fragment of the mineral on char- 
 coal, or with a forceps, in the oxidising point of the blowpipe 
 flame. 
 
 1. Fumes or odour given off. (See Group IV.) 
 
 2. No fumes or odour given off. (See Group V.) 
 
 B. The streak is uncoloured that is to say, white, or 
 nearly so. 
 
 Experiment. Try to scratch a smooth surface of the 
 mineral with a splinter of quartz. 
 
 3. The mineral can be scratched by quartz. (See 
 Group VI.) 
 
 4. The mineral cannot be scratched bv quartz. (See 
 Group VII.) 
 
 Note 1. The determination must always be made upon pure minerals 
 free from any adhering foreign matter. 
 
 Note 2. Coal and carbonaceous shale are distinguished from minerals 
 which they may resemble by burning in the blowpipe flame, leaving a more 
 or less copious ash. 
 
 GROUP I. 
 
 MINERALS WHICH HAVE A METALLIC LUSTRE, AND WHICH GIVE OFF 
 
 SULPHUR. 
 
 Experiment. Try to scratch the mineral with a knife, 
 and observe the streak. 
 
 I. Not scratched with a knife ; scratches glass witli 
 great facility. 
 
 Pale brass-yellow or white. Iron pyrites , 41. 
 
 II. Scratched with a knife ; does not scratch glass 
 easily. 
 
 * Streak unmetallic. 
 
282 DETERMINATION OF MINERALS 
 
 a. Brass-yellow. Solution in nitric acid blue. Copper 
 pyrites, 66. 
 
 b. Black or red. Solution in nitric acid colourless, but 
 on adding salt water a white curd of precipitate is thrown 
 down which blackens on exposure to sunlight. Mineral 
 gives off white fumes before the blowpipe. Antimonial or 
 arsenical silver ores, 75. 
 
 c. Lead- or steel-grey. Abundant white fumes before 
 the blowpipe. On charcoal with sodium carbonate yields 
 a globule of metallic lead. Lead and antimony sulphides, 77. 
 
 d. Eed. Wholly volatile before the blowpipe. Mixed 
 with carbonate of soda and heated in an iron spoon over a 
 candle flame, white vapours are given off which may be 
 condensed on a gold coin held a little above the mixture, 
 covering its surface with a brilliant amalgam when rubbed. 
 Cinnabar i 62. 
 
 Note. The streak of Grey Copper (see below) may be unmetallic, often 
 red. 
 
 ** Streak metallic. 
 
 Experiment. Try fusibility in bio wpipe flame. 
 
 A. Infusible : 
 
 e. Lead- grey. Very soft. In thin leaves. Tinges the 
 blowpipe flame faint green. Molybdenite, 35. 
 
 Note. Magnetic Iron Pyrites and Vitreous Copper are fusible with diffi- 
 culty. 
 
 B. Fusible : (Easily, except k and m.) 
 
 Experiment. Heat a small piece (size of linseed), free 
 from gangue, on charcoal without adding sodium carbonate,, 
 hold first in oxidising point and finish in reducing point. 
 
 f. Wholly volatile, emitting abundant white fumes. 
 Mineral has a lead-colour. Crystallised in long prisms, 
 Antimony sulphide, 37. 
 
 g. Globule of metallic lead remains. Mineral has a 
 lead-colour. Cubical cleavage. Galena, 58. 
 
 h. Abundant white fumes given off. Mineral has a lead 
 or steel colour ; and when heated on charcoal with sodium 
 carbonate affords a globule of metallic lead. Lead and 
 antimony sulphides, 77. 
 
BEFORE THE BLOWPIPE. 283 
 
 i. Globule of metallic silver remains (facilitated by 
 adding a little sodium carbonate). Mineral has a black 
 colour. Is soluble in nitric acid ; on adding salt water 
 to the solution a white curd is thrown down which blackens 
 on exposure to sunlight. Silver sulphide, 74. 
 
 k. Globule of metallic copper remains (facilitated by 
 adding a little sodium carbonate) ; discovered by crushing 
 the residue and flattening it under a hammer. Mineral ha? 
 a black colour. Does not give off copious white fumes 
 before the blowpipe. It forms a blue solution with nitric 
 acid. Vitreous copper, 65. 
 
 I. Slaggy globule remains, usually attracted by a 
 magnet, not yielding copper when crushed. Mineral has 
 a black or steel colour. It gives off copious white fumes 
 before the blowpipe. It forms a greenish-brown solution 
 with nitric acid. Grey copper + 67. 
 
 m. Greyish-black magnetic bead remains. Mineral 
 has a bronze or copper colour. Can be scratched with a 
 knife ; is attracted by a magnet ; does not form a blue or 
 green solution with nitric acid. Magnetic iron pyrites, 85. 
 
 n. Black magnetic globule. Mineral has the form of 
 brass yellow needles. Nickel sulphide, 79. 
 
 N.B. Before determining the name of a mineral, it is necessary to com- 
 pare it with the description to which the number refers. 
 
 GBOUP II. 
 
 MINERALS WHICH HAVE A METALLIC LUSTRE AND WHICH GIVE OFF 
 EITHER AN ODOUR OF GARLIC WITHOUT SULPHUR, OR WHITE FUMES 
 WHICH HAVE NOT A GARLIC OR SULPHUROUS ODOUR. 
 
 1. The 1 fumes evolved have an odour of garlic 
 (arsenic). 
 
 * Not scratched with a knife. Scratches glass. 
 
 a. White metallic. Eesidue after roasting gives indi- 
 cations of iron, with borax bead. Crystallised in flattened 
 prisms. Mispickel, 42. 
 
 b. White metallic. Eesidue after roasting gives indi- 
 
284 DETERMINATION OF MINERALS 
 
 cations of cobalt, with borax bead. Crystallised in octa- 
 hedrons. Smaltine, 53. 
 
 Note. White Nickel resembles Smaltine, and is found with it. 
 ** Scratched with a knife. Does not scratch glass. 
 
 c. Pale copper red. Arsenical nickel, 52. 
 
 d. Carmine red. Arsenical silver ore, 75. 
 
 e. White (as streak shows), tarnished black ; wholly 
 volatilised by heat. Arsenic, 38. 
 
 Note. Metallic Bismuth is frequently associated with Arsenic. Possibly 
 Grey Copper and Arsenical Silver may be looked for here ; for their charac- 
 ters see Group I. 
 
 2. The furnes evolved have not a garlic odour. 
 
 Possibly grey copper, antimonial silver ore, lead and 
 antimony sulphides, or cinnabar : but these ores generally 
 give off a faint odour of sulphur in addition to white 
 fumes. For their characters, see Group I. 
 
 N.B. Before determining the name of a mineral, it is necessary to com- 
 pare it with the description to which the number refers. 
 
 GROUP III. 
 
 MINERALS WHICH HAVE A METALLIC LUSTRE, AND WHICH GIVE OFF 
 
 NO FUMES. 
 
 I. The mineral is malleable (can be flattened under a 
 hammer). 
 
 a. Yellow. Gold, 72. 
 
 b. Eed. Copper, 64. 
 
 c. White, rusty surface ; strongly attracted by a 
 magnet. Iron, 40. 
 
 d. White, feebly or not attracted by a magnet. In- 
 fusible. Insoluble in nitric acid. Platinum, 71. 
 
 e. White, often tarnished ; not attracted by a magnet. 
 Fusible. Soluble in nitric acid ; on adding salt water to 
 the solution, a white curd is thrown down, which blackens 
 on exposure to sunlight. Silver, 73. 
 
 II. The mineral is brittle (breaks to pieces under a 
 hammer). 
 
BEFORE THE BLOWPIPE. 285 
 
 Experiment. Observe the colour imparted to a borax 
 bead in oxidising and in reducing. 
 
 1. Violet in oxidising, colourless in reducing. 
 
 a. Manganese ores, 51. 
 
 2. Red (hot), yellow (cold), in oxidising, bottle-green in 
 reducing. 
 
 Scratch the mineral with quartz and observe the colour of the streak. 
 
 b. Brown streak ; mineral not attracted by a magnet. 
 Brown iron oxide, 46. 
 
 c. Eed streak ; not attracted by a magnet. Specular 
 iron, 44. 
 
 d. Black streak ; not attracted by a magnet. Titanic 
 iron, 47. 
 
 e. Black or brown streak ; strongly attracted by a 
 magnet. Magnetic iron, 43. 
 
 Note. Possibly Wolfram (83) ; non -magnetic ; streak reddish-brown or 
 black ; remarkable for its great weight. 
 
 3. Green (hot], blue (cold), in oxidising, m^in reducing, 
 Red, cuprous oxide, 69. 
 
 4. Green in oxidising, green in reducing. 
 /. Chromic iron, 48. 
 
 5. Colourless. 
 
 * Infusible. 
 
 Minerals HARDER than calc-spar. 
 
 g. Eeadily soluble in nitric acid. Scratched with a 
 knife. Blende, 55. 
 
 h. Insoluble in nitric acid. Not scratched with a 
 knife. With potassium cyanide upon charcoal, yields a 
 globule of metallic tin. Tin ore, 34. 
 
 Minerals SOFTER than calc-spar. 
 
 i. Black, like black-lead. Specific gravity about 2. 
 Graphite, 20. 
 
286 DETERMINATION OF MINERALS 
 
 k. Lead-colour. Specific gravity about 4:5. linger 
 blowpipe flame pale green. Molybdenite, 
 
 ** Fusible. 
 
 35. 
 
 Heated on charcoal forms an incrustation which is red 
 whilst hot, yellow when cold. Bismuth, 36. 
 
 6. Note. Green in oxidising, dirty violet in reducing, indicates Entile 
 (81) ; yellow in oxidising, green in reducing (difficult to obtain in a candle 
 flame), afforded by a black mineral, soluble in boiling nitric acid, forming a 
 yellow solution, indicates Pitchblende (84). 
 
 Note. Mica may possibly be sought for in this group. It is distinguished 
 by its lightness of weight, and its capability of being split into very thin, trans- 
 parent, flexible layers. 
 
 N.B. Before determining the name of a mineral, it is necessary to com- 
 pare it with the description to which the number refers. 
 
 GROUP IV. 
 
 MINERALS WHICH POSSESS AN UNMETALLIC LUSTRE, A COLOURED STREAK, 
 AND WHICH GIVE OFF FUMES OR OB-OUR WHEN HEATED BEFORE THE 
 BLOWPIPE. 
 
 a. Colour and streak yellow, entirely volatile, burning 
 with a blue flame and sulphurous odour. Sulphur, 33. 
 
 b. Colour red, orange, or yellow, entirely volatile, burns 
 on charcoal with a blue flame and garlic odour. Arsenic 
 sulphide, 39. 
 
 c. Streak red. Strong odour of garlic or white fumes 
 (of antimony) when heated. Dissolved in nitric acid ; on 
 adding salt to the solution a white curd is thrown down 
 which blackens on exposure to sunlight. Arsenical or 
 antimonial silver ore, 75. 
 
 d. Streak red. Wholly volatile before blowpipe. 
 Mixed with sodium carbonate and heated in an iron spoon 
 over a candle flame, vapours are given off which may be 
 condensed on a gold coin held a little above the mixture, 
 covering its surface with a brilliant amalgam when rubbed. 
 Cinnabar, 62. 
 
 Sfjj. Before determining the name of a mineral, it is necessary to com- 
 pare it with the description to which the number refers. 
 
BEFORE THE BLOWPIPE. 287 
 
 GEOUP V. 
 
 MINERALS WHICH POSSESS AN UNMETALLIC LUSTKE, A COLOURED STREAK, 
 BUT WHICH GIVE OFF NO FUMES OR ODOUR BEFORE THE BLOWPIPE. 
 
 Experiment Observe the colour imparted by the mineral 
 to a borax bead, both in oxidising and reducing. 
 
 1. Green (hot), blue (cold), in oxidising ; red in reduc- 
 ing (difficult to obtain). 
 
 a. Colour of mineral blue or green. Copper carbonate , 
 70. 
 
 b. Colour of mineral red. Red cuprous oxide , 69. 
 
 c. Colour of mineral black. Black cuprous oxide, 68. 
 
 2. Green in oxidising, green in reducing. 
 
 d. Colour of mineral black. Chromic iron, 48. 
 
 3. Red (hot), yellow (cold), in oxidising : bottle-green in 
 reducing. 
 
 e. Colour of mineral green. Greenearth, 40. 
 
 /. Colour of mineral brown ; blackened during heating. 
 Brown ferric oxide, 46. 
 
 g. Colour of mineral red ; blackened during heating. 
 Red ferric oxide, 45. 
 
 Note. Iron Arseniate is green or yellow, Iron Phosphate is blue. 
 
 4. Amethyst in oxidising, colourless in reducing. 
 
 h. Colour of mineral brown or black. Manganese 
 -ores, 51. 
 
 Note. Manganese Carbonate is rose red or brownish, streak white. 
 
 5. Colourless in both oxidising and reducing. 
 
 i. Easily scratched with a knife. Cleavage. Soluble 
 in nitric acid. Blende, 55. 
 
 k. Not scratched with a knife. No cleavage. Heated 
 with potassium cyanide upon charcoal yields a malleable 
 globule of metallic tin. Tin ores, 34. 
 
 Note. Cobalt Bloom (54) has a red colour and streak ; it gives the indi- 
 cations of Cobalt with borax bead. Entile, Wolfram, and Pitchblende might 
 be considered to have an unmetallic lustre. See under Group III. 
 
 Njg. Before determining the name of a mineral, it is necessary to com- 
 pare it with the description to which the number refers. 
 
283 DETERMINATION OF MINERALS 
 
 GEOUP VI. 
 
 MINERALS WHICH HAVE AN UNMETALLIC LUSTRE, AND WHICH ARE 
 SCRATCHED BY QUARTZ, SHOWING A WHITE STREAK. 
 
 I. Minerals soluble in water (having a taste). 
 
 a. Taste of common table salt. Salt, 28. 
 
 b. Sweetish astringent taste. Alum, 29. 
 
 c. Bitter taste of Epsom salts. Epsomite, 29. 
 
 d. Cooling taste. Causes vivid combustion when 
 thrown on a piece of red-hot charcoal. Nitre, 30. 
 
 Note. Zinc (white), Iron (green), and Copper (blue) sulphates have a 
 nauseous metallic taste. 
 
 II. Minerals insoluble in water ; but soluble in nitric 
 acid (best determined by placing a very little of the 
 powder in a test tube, pouring on it a few drops of the 
 acid, and heating if requisite). 
 
 N.B. All of these can be scratched witli a knife. 
 * Easily fusible. 
 
 a. Green, yellow, or brown. Hardness, 3 to 4. Glo- 
 bule of lead, on charcoal with sodium carbonate. No 
 effervescence in nitric acid. Pyromorphite, 61. 
 
 b. White or grey. Hardness, 3 to 4. Flies violently 
 to pieces in the blowpipe flame. Globule of lead on char- 
 coal with sodium carbonate. Effervesces in nitric acid. 
 Cerusite, 59. 
 
 Note. White or grey. Effervesces in nitric acid. Specific gravity 4'3. 
 Does not fly to pieces when heated ; yields no metal on charcoal, WITHERITE 
 
 BARIUM CARBONATE. 
 
 ** Infusible. 
 
 c. Hardness, 3. Dissolves with very brisk efferves- 
 cence in cold nitric acid. Does not crumble to powder in 
 the blowpipe flame. Calc-spar, 24. 
 
 d. Hardness, 3- 5 to 4. Dissolves like calc-spar ; but 
 crumbles to powder in the blowpipe flame. Aragonite, 6. 
 
 e. Hardness, 3 -5 to 4. Dissolves much slower than 
 calc-spar ; but difficult to distinguish without chemical 
 analysis. Dolomite, 26. 
 
BEFOEE THE BLOWPIPE. 289 
 
 f. Hardness, 4 to 5. No effervescence in cold nitric 
 acid, very slight in hot. Magnesite, 25. 
 
 g. Hardness, 3'5 to 4*5. Effervescence in hot nitric 
 acid. The mineral blackens when heated in the blowpipe 
 flame ; it gives the indication of iron with borax. Iron 
 carbonate, 50. 
 
 h. Hardness, 5. Effervescence in hot nitric acid. 
 Heated on charcoal w r ith sodium carbonate it forms a 
 crust which is yellow whilst hot. Ckdamine, 56. 
 
 i. Hardness, 5. Dissolves in nitric acid without effer- 
 vescence, leaving a jelly of silica. In blowpipe flame it 
 shines with a green light. Zinc silicate. 57. 
 
 j. Hardness, 3-5. Effervesces in hot nitric acid. Flies 
 violently to pieces in the blowpipe flame. On charcoal 
 with sodium carbonate yields a globule of lead. Cerusite, 
 59. 
 
 Jc. Hardness, 3- 5 to 4. Brilliant cleavage. Eapidly 
 soluble in nitric acid with effervescence. Heavy (specific 
 gravity, 4). Blende, 55. 
 
 1. Hardness, 4- 5 to 5. Slowly soluble in nitric acid 
 without effervescence. Apatite, 22. 
 
 Note. Hardness, 3'5. Effervesces in hot nitric acid. Tinges the blow- 
 pipe flame crimson. Strontianite (Strontium carbonate). 
 
 HI. Minerals insoluble in both water and nitric acid. 
 
 * Easily fusible. 
 
 HARDNESS. 
 
 TO to 1*5. Like wax. On charcoal with sodium carbon- 
 ate yields a globule of silver. Horn silver, 
 
 76. 
 2-5 to 3-0. White or grey. On charcoal with sodium 
 
 carbonate yields a globule of lead. Angle- 
 site, 60. 
 3-5 to 4-0. Green or brown. On charcoal with sodium 
 
 carbonate yields a globule of lead. Pyro- 
 
 morphite, 61. 
 3*5 to 6'5. Usually white. Swell up before the blowpipe. 
 
 Occur in cavities of rocks. Zeolites, 15. 
 5-0 to 6-0. Black, green, or white. Prisms, fibrous or 
 
 flax-like. Hornblende and augite, 6. 
 
 u 
 
290 DETERMINATION OF MINERALS 
 
 HARDNESS. 
 
 6 -5 to 7 *5. Usually red. Crystallised in dodecahedrons. 
 Garnet, 9. 
 
 Note. Glass (artificial) may be included here ; easily distin- 
 guished by its appearance, easy fusibility, and inferior hard 
 ness, 4*5 to 5*5. 
 
 ** Infusible, or not easily fusible. 
 
 rO to 1-5. Light green. In layers. Unctuous. Talc, 3. 
 1-5. Dark green. In layers or granular. Some- 
 times fusible to a black glassy bead. Chlo- 
 rite , 4. 
 
 2'0 to 3*5. Green. Turns white in blowpipe flame. Feels 
 
 often greasy. Serpentine, 5. 
 
 2-0. Usually white. After heating becomes opaque, 
 and can be rubbed to powder between the 
 fingers. Fusible with difficulty. Gypsum, 31. 
 
 2-0 to 2-5. Grey or black. Can be split into extremely 
 thin semi-transparent flexible layers. Mica, 
 14. 
 
 3-0 to 3-5. White or grey. Tabular crystals or massive. 
 Heavy (specific gravity, 4-3 to 4-8). Splinters 
 fly off crystals when heated. Fusible with 
 difficulty. Heavy spar, 32. 
 
 4-0. Cubes or massive. Octahedral cleavage. Flies 
 to pieces when heated. Specific gravity, 3 
 to 3-2. Fusible with difficulty. Fluor-spar, 
 23. 
 
 5-0. Six-sided prisms, massive. Green or white. 
 Specific gravity, 3 to 3 -3. Generally flies 
 to pieces when heated. In fine powder, 
 slowly soluble in nitric acid. Apatite, 22. 
 
 5-0 to 6-0. Prisms, fibrous or flax-like. Cleavage. Green, 
 black, or white. Generally fusible. Specific 
 gravity, 2-9 to 3-5. Hornblende and augite, 6. 
 
 Note. Possibly Epidote. Yellowish green. Hardness, 6'0 
 to 7-0. 
 
 6-0. Prisms or massive. Flesh-colour, white, or 
 tinted. Cleavage. Specific gravity, 2-3 to 
 2-8. Felspar, 13. 
 
BEFORE THE BLOWPIPE. 291 
 
 HARDNESS. 
 
 () to 7*0. Masses or grains like glass. Black or green. 
 Specific gravity, 3'3 to 3'6. Chrysolite, 7. 
 
 6-5 to 7' 5. Dodecahedrons. Usually red. Generally fus- 
 ible. Eed infusible varieties impart a green 
 colour to borax bead. Specific gravity, 3*5 
 to 4-3. Garnet, 9. 
 
 5-5 to 6'5. Never crystallised. White, tinted, or chatoyant. 
 
 Specific gravity, 1-9 to 2-3. Opal, 2. 
 7'0. Six-sided prisms. Specific gravity, 2 -6. 
 Quartz, 1. 
 
 Note. Spliene (82) occurs in acute thin crystals. Yellow, 
 green, or brown. Specific gravity, 3'2 to 3'6. 
 
 Chiastolite and other Aluminium Silicates occur in 
 prisms. Brown or white. Opaque. Hardness, 6 to 7'5. 
 Infusible. Specific gravity, 3 to 3*7. 
 
 N.B. Before determining the name of a mineral, it is neces- 
 sary to compare it with the description to which the num- 
 ber refers. 
 
 GROUP VIL Including the Gems. 
 
 MINERALS WHICH HAVE AN UNMETALLIC LUSTRE, AND WHICH ARE 
 NOT SCRATCHED BY QUARTZ. 
 
 Note. Tin ore (34) may occur here. Hardness, 6 to 7. Bemarkable for 
 its great weight. Specific gravity, 6'8 to 7'0. Brown or black. With cya- 
 nide of potassium, on charcoal, yields a globule of Tin. 
 
 GEMS. 
 
 If crystallised, the determination is greatly facilitated, 
 but if the crystallisation is not evident, the hardness will 
 be found a sufficiently near indication in the following 
 scheme. The requisites in determining the hardness of 
 gems are good specimens of sapphire, topaz, and quartz 
 possessing smooth bright surfaces, as well as sharp points 
 or corners. The hardness is determined by drawing the 
 sharp points of these three test-minerals over smooth bright 
 surfaces of the mineral under trial. (See hardness.) Or 
 it may be ascertained with great certainty by means of a 
 small angular fragment broken off the mineral to be tried. 
 For convenience of holding, it should be mounted on a 
 stick of sealing-wax, which is best done by previously 
 heating the fragment, held with a forceps over a candle 
 flame, and applying it, whilst hot, to the wax ; the hard 
 
 U 2 
 
292 DETERMINATION OF MINERALS 
 
 ness may then be ascertained by drawing the fragment 
 thus mounted across smooth surfaces of the sapphire, 
 topaz, and quartz, and observing whether a scratch is 
 thus produced. In this way the hardness of a mineral not 
 larger than a grain of sand may be determined. 
 
 The facility of cleavage is an important character in 
 the topaz and the diamond. 
 
 The electrical properties are characteristic in the dia- 
 mond, topaz, and tourmaline. 
 
 The specific gravity is a most important aid in distin- 
 guishing gems, but it cannot be attempted with an ordinary 
 pair of grain scales in the case of minerals weighing under 
 ten grains at the very least. 
 
 Many varieties of zircon and spinel, and in some cases 
 the topaz and emerald, can readily be distinguished by the 
 effect of heat, which is applied most conveniently by the 
 blowpipe. 
 
 The diamond may be easily distinguished by the use of 
 a small writing or scratching diamond, which fails to mark 
 the faces of a real diamond when drawn lightly across, but 
 scratches all other gems with facility. After practice, the 
 sound caused by tapping two diamonds together becomes 
 characteristic. 
 
 1. CRYSTALLISED IN PRISMS. 
 
 HARDNESS. 
 
 7*0. Six-sided prisms or pyramids ; sides of prism 
 finely marked across. No cleavage. Frac- 
 ture fconchoidal. Infusible. When two 
 pieces are Drubbed together in the dark, 
 they emit a phosphorescent light and a 
 peculiar odour. Specific gravity, 2 -5 to 2*8. 
 Quartz, 1. 
 
 7*0 to 8*0. Prisms, three, six, nine, or more sided ; fur- 
 rowed lengthwise. Black or coloured. 
 Opaque or transparent. Smooth sides of 
 prisms become electric by friction. No 
 cleavage. Infusible or nearly so. Specific 
 gravity, 3'0 to 3- 3. Tourmaline, 8. 
 
BEFORE THE BLOWPIPE. 293 
 
 FARDNESS. 
 
 7*5. Prisms, four-sided, not furrowed. Wine-red, 
 brown, or white. Decolourised permanently 
 by heat. Adamantine cleavage, but rather 
 difficult to obtain. Infusible. Specific 
 gravity, 4 to 5. Zircon, 12, 
 
 7 to 8. Prisms, six or more sided. Usually green. 
 Imperfect cleavage. Fusible with difficulty 
 in thin edges ; becoming clouded by heat. 
 Specific gravity, 2'7 ; its low specific gravity 
 combined with its hardness is characteristic. 
 Beryl or Emerald, 11. 
 
 7'5 to 8. Sides of prisms often finely marked lengthwise. 
 Cleavage across the prisms brilliant and 
 easily obtained. Smooth surfaces become 
 strongly electric by friction. Infusible, but 
 sometimes blistered and altered in colour by 
 heat. Specific gravity, 3-5. Topaz, 10. 
 8'5. Prisms or tables. Green. Infusible and un- 
 altered by heat. Specific gravity, 3-5 to 3 '8. 
 Chrysoberyl, 18. 
 
 9'0. Prisms, six-sided or irregular. Cleavage across 
 the prisms, but difficult to obtain. Fracture 
 irregular. Blue, green, black, red, yellow, 
 brown, or white. Infusible. Easily distin- 
 guished by its great hardness, scratching 
 all other gems except diamond. Specific 
 gravity, 3*9 to 4- 2. Sapphire, ruby, or co- 
 rundum, 16. 
 
 2. CRYSTALLISED IN OCTAHEDRONS OR DODECAHEDRONS. 
 
 HARDNESS. 
 
 7'5. See characters above. Zircon, 12. 
 '6-5 to 7-5. Dodecahedrons. Usually red. Most varieties 
 are easily fusible ; red infusible varieties 
 impart a green colour (due to chromium) 
 to borax bead. Specific gravity, 3-1 to 4-3. 
 Garnet, 9. 
 
294 
 
 COLOURED FLAMES. 
 
 HARDNESS. 
 
 8-0. Octahedrons. Usually red or black, also blue,, 
 green, yellow, and brown. Some red varie- 
 ties become opaque- and black when heated ; 
 rose-red varieties become green when heated,, 
 but regain their original colour on cooling. 
 Specific gravity, 3 -5 to 4'0. Spinel, 17. 
 10-0. Octahedrons, dodecahedrons, or modifications 
 of these forms. Crystalline facets often 
 curved. Cleavage perfect. Lustre, brilliant 
 adamantine. Usually colourless or straw- 
 coloured. Not water-worn. Strongly elec- 
 tric by the slightest friction. Specific gravity, 
 3-5. Diamond, 19. 
 
 N.B. Before determining the name of a mineral, it is neces- 
 sary to compare it with the description to which the number 
 refers. 
 
 For further particulars respecting gems and precious 
 stones see the concluding chapter of this volume. 
 
 COLOURED FLAMES. There are a great number of sub- 
 stances best detected by the colours they impart to the 
 flame of the blowpipe. Indeed, so important is this point 
 that it has been thought advisable to collect all the facts 
 known on this subject into one place, rather than scatter 
 them over the work. These experiments are best made 
 in a dark room, and with a very small flame.* 
 
 BLUE FLAMES. 
 
 Large intense blue . Copper chloride. 
 
 Pale clear blue . . Lead. 
 
 Light blue . . Arsenic. 
 
 Blue ... . Selenium. 
 
 Greenish blue . . Antimony. 
 
 Blue mixed with green Copper bromide. 
 
 GREEN FLAMES. 
 
 Intense emerald green . Thallium. 
 
 Very dark green, feeble . Ammonia. 
 
 Dark green . . . Boracic acid. 
 
 Full green . . . Tellurium. 
 
 Full green Copper. 
 
 Emerald green mixed with blue . \ PP er j did ?- 
 
 ( Copper bromide. 
 
 Pale green . . . . Phosphoric acid. 
 
 Very pale apple -green . . . Barium. 
 
 Intense whitish green . . . Zinc. 
 
 Bluish green Tin binoxide. 
 
 * Griffin's ' Blowpipe Analysis,' p. 148. 
 
COLOURED FLAMES. 295 
 
 YELLOW FLAME. 
 
 Intense greenish yellow . . . Sodium. 
 
 RED FLAMES. 
 
 Intense crimson .... Lithium. 
 
 Eed Strontium. 
 
 Keddish purple Calcium. 
 
 Violet Potassium. 
 
 Chlorine, combined with copper, gives an intense blue 
 flame. This phenomenon may be produced as follows : 
 Take a piece of thin brass wire, and bend one end of it 
 several times upon itself; place upon this some microcosmic 
 salt, and fuse it until it has acquired a green colour. Then 
 add to it the substance suspected to contain chlorine, and 
 place it in the oxidising flame, just at the point of the blue 
 flame ; if any chloride be present a splendid blue colour 
 will be produced. 
 
 Lead. The blue colour produced by this metal is 
 readily obtained. Fragments of a mineral must be held 
 in the tongs, and powder may be assayed on charcoal. 
 
 Arsenic, in the metallic state, gives rise to a light blue 
 flame. 
 
 Selenium and Antimony, when treated in the same 
 manner, afford characteristic flames. 
 
 Bromine. If any substance containing bromine be 
 placed in a bead of fused microcosmic salt on the brass 
 wire, and then in the oxidising flame, a bright blue flame 
 with emerald-green edges will be produced. 
 
 Boracic Acid. The following is Dr. Turner's process 
 for the detection of boracic acid. The substance is to be 
 mixed with a flux composed of 1 part of fluor-spar and 4J 
 parts of potassium bisulphate. This mixture is to be 
 made to adhere to the moistened end of a platinum wire, 
 and held at the point of the blue flame ; at the instant of 
 fusion a dark green flame will be produced. It may also 
 be produced by merely dipping the mineral in sulphuric 
 acid, and exposing it to the blowpipe blast. In case a 
 very small quantity of boracic acid is contained in a 
 mineral, the following process may be employed : The 
 substance must be fused with potassium carbonate on 
 
296 COLOURED FLAMES. 
 
 charcoal, moistened with a drop or two of sulphuric acid, 
 and then a few drops of alcohol ; the latter will burn 
 with a green flame when exposed to the flame of the 
 blowpipe. 
 
 Tellurium. The peculiar flame given by this metal is 
 produced by heating a portion of its oxide on charcoal in 
 the reducing flame. 
 
 Copper. All the compounds of copper, except those 
 in which bromine and chlorine enter, give a beautiful 
 green flame. The soluble salts give it per se, but the 
 insoluble require moistening with sulphuric acid. 
 
 Iodine and Copper. To the bead of microcosmic salt 
 on the brass wire add any compound containing iodine, 
 and a bright green flame will be produced when the mass 
 is heated in the oxidising flame. 
 
 Phosphoric Acid. The phosphates, when moistened 
 with sulphuric acid, give a light green tint to the outer 
 flame. 
 
 Barium. The soluble salts of barium give a light green 
 colour to the outer flame when moistened with water. 
 
 Zinc, when exposed to the blowpipe flame, burns with 
 an intense whitish-green light. 
 
 Sodium. Any salt of soda, or substance containing soda, 
 being exposed to the outer flame, gives a brush of intensely 
 coloured flame, of a fine amber or greenish-yellow. 
 
 Water. Certain minerals containing water give a feeble 
 yellowish tint to the flame due to the presence of a little 
 sodium. 
 
 Strontium. All the salts of this substance which are 
 soluble in water give a crimson tint to the flame, which 
 does not endure after the substance is fused. Strontium 
 carbonate must be moistened with hydrochloric acid, and 
 strontium sulphate must be reduced to a state of sulphide 
 by ignition with charcoal ; it must then be moistened with 
 hydrochloric acid, after which treatment it will exhibit 
 the characteristic flame. 
 
 Lithium. All that has been said of strontium applies to 
 lithium, but the coloured flame given by lithium is perma- 
 nent, whilst that afforded by strontium is evanescent. 
 
COLOURED FLAMES, 297 
 
 Calcium acts as strontium. 
 
 Potassium, treated as sodium, gives a purplish light. 
 This is, however, very liable to be masked by the intense 
 yellow communicated by small quantities of sodium, which 
 are almost always present with potassium. The potassium 
 flame can be seen even in presence of sodium by looking 
 through cobalt-coloured glass. 
 
CHAPTER VIII. 
 
 VOLUMETRIC ANALYSIS. 
 
 THE main feature of volumetry is not so much analysis in 
 the proper sense of the term, as the quantitative estima- 
 tion of one principal constituent of a substance. 
 
 This estimation is done by means of solutions, con- 
 taining a certain quantity of reagents in a certain volume, 
 which is called a standard solution, the quantity used of 
 such solution being measured by graduated tubes (burettes, 
 pipettes, &c.) 
 
 The reaction of a volumetric analysis can be of three 
 different kinds, according to the reagent used and to the 
 substance to be estimated. 
 
 1. The substance to be analysed being an acid or a 
 base, it can be saturated by a suitable standard solution 
 (saturation-analysis, used for acids, potash, soda, &c.) 
 
 2. The substance to be assayed maybe precipitated by 
 the standard solution, and the completion of the process is 
 observed when no further precipitate occurs (precipitation- 
 analysis, e.g. Pelouze's copper assay, Gay-Lussac's silver 
 assay). 
 
 3. The substance to be estimated becomes, by the 
 standard solution, either oxidised or reduced, and by the 
 performance of this process certain colours will appear or 
 disappear, from which the completion of the process is to 
 be observed (oxidation or reduction-analysis, e.g. Schwarz's 
 copper assay). 
 
 These processes of volumetric analysis are frequently 
 used in assaying. 
 
 The principle of volumetric analysis may be fully ex~ 
 plained by the following examples given by Fresenius.* 
 
 * Fresenius's ' Quantitative Analysis,' fourth edition, p. 76. 
 
PEINCIPLES O'F VOLUMETRIC ANALYSIS. 299 
 
 ' Suppose we have prepared a solution of chloride of 
 sodium of such a strength that 100 c.c. will exactly pre- 
 cipitate 1 grm. silver from its solution in nitric acid, we 
 can use it to estimate unknown quantities of silver. Let 
 us imagine, for instance, we have an alloy of silver and 
 copper in unknown proportion, we dissolve 1 grm. in nitric 
 acid, and add to the solution our solution of chloride of 
 sodium, drop by drop, until the whole of the silver is 
 thrown down, and an additional drop fails to produce a 
 further precipitate. The amount of silver present may 
 now be calculated from the amount of solution of chloride 
 of sodium used. Thus, supposing we have used 80 c.c., 
 the amount of silver present in the alloy is 80 per cent. ; 
 since, as 100 c.c. of the solution of chloride of sodium will 
 throw down 1 grm. of pure silver (i.e. 100 per cent.), it 
 follows that every c.c. of the chloride of sodium solution 
 corresponds to 1 per cent, of silver. 
 
 ' Another example. It is well known that iodine and 
 sulphuretted hydrogen cannot exist together ; whenever 
 these two substances are brought in contact decomposi- 
 tion immediately ensues, the hydrogen separating from the 
 sulphur and combining with the iodine (I + HS = HI + S). 
 Hydriodic acid exercises no action on starch paste, whereas 
 the least trace of free iodine colours it blue. Now, if we 
 prepare a solution of iodine (in iodide of potassium) con- 
 taining in 100 c.c. 0-7470 grm. iodine, we may with this 
 decompose exactly O'l grm. sulphuretted hydrogen ; for 
 17 : 127:: 0-1 : 0-7470. 
 
 ' Let us suppose, then, we have before us a fluid con- 
 taining an unknown amount of sulphuretted hydrogen, 
 which it is our intention to estimate. We add to it a 
 little starch paste, and then, drop by drop, our solution 
 of iodine, until a persistent blue coloration of the fluid 
 indicates the formation of iodide of starch, and hence the 
 complete decomposition of the sulphuretted hydrogen. 
 The amount of the latter originally present in the fluid 
 may now be readily calculated from the amount of solution 
 of iodine used. Say, for instance, we have used 50 c.c. 
 of iodine solution, the fluid contained originally 0-05 sul- 
 
300 PRINCIPLES OF VOLUMETRIC ANALYSIS. 
 
 phuretted hydrogen ; since, as we have seen, 100 c.c. of 
 our iodine solution will decompose exactly O'l grm. of 
 that body. 
 
 ' Solutions of accurately known composition or strength, 
 used for the purposes of volumetric analysis, are called 
 standard solutions. They may be prepared in two ways, 
 viz. (a) by dissolving a weighed quantity of a substance 
 in a definite volume of fluid ; or (b) by first preparing a 
 suitably concentrated solution of the reagent required, 
 and then estimating its exact strength by a series of 
 experiments made with it upon weighed quantities of the 
 body for the estimation of which it is intended to be 
 used. 
 
 ' In the preparation of standard solutions by method a, 
 a certain definite strength is adopted once for all, which is 
 usually based upon the principle of an exact correspond- 
 ence between the number of grammes of the reagent 
 contained in a litre of the fluid, and the equivalent number 
 of the reagent (H=l). In the case of standard solutions 
 prepared by method b, this may also be easily done, by 
 diluting to the required degree the still somewhat too 
 concentrated solution, after having accurately estimated 
 its strength ; however, as a rule, this latter process is only 
 resorted to in technical analyses, where it is desirable to 
 avoid all calculation. Fluids which contain the equivalent 
 number of grammes of a substance in 1 litre are called 
 normal solutions ; those which contain y 1 ^ of this quantity, 
 decinormal solutions. 
 
 ' The estimation of a standard solution intended to 
 be used for volumetric analysis is obviously a most im- 
 portant operation ; since any error in this will, of course, 
 necessarily falsify every analysis made with it. In scien- 
 tific and accurate researches it is, therefore, always ad- 
 visable, whenever practicable, to examine the standard 
 solution no matter whether prepared by method a or 
 by method />, with subsequent dilution to the required 
 degree by experim en ting with it upon accurately weighed 
 quantities of the body for the estimation of which it is 
 to be used. 
 
PRINCIPLES OF VOLUMETRIC ANALYSIS. 301 
 
 ' In the previous remarks no difference has been made 
 between fluids of known composition and those of known 
 power ; and this has hitherto been usual. But by accept- 
 ing the two expressions as synonymous, we take for granted 
 that a fluid exercises a chemical action exactly correspond- 
 ing to the amount of dissolved substance it contains ; that, 
 for instance, a solution of chloride of sodium containing 
 1 eq. NaCl will precipitate exactly 1 eq. silver. This pre- 
 sumption, however, is very often not absolutely correct. 
 In such cases, of course, it is not merely advisable, but even 
 absolutely necessary, to estimate the strength of the fluid 
 by experiment, although the amount of the reagent it 
 contains may be exactly known, for the power of the fluid 
 can be inferred from its composition only approximately,, 
 and not with perfect exactness. If a standard solution 
 keeps unaltered, this is a great advantage, as it dispenses 
 with the necessity of estimating its strength before every 
 fresh analysis. 
 
 ' That particular change in the fluid operated upon by 
 means of a standard solution, which marks the completion 
 of the intended decomposition, is termed the FINAL REACTION. 
 This consists either in a change of colour ', as is the case 
 when a solution of permanganate of potash acts upon an 
 acidified solution of protoxide of iron, or a solution of 
 iodine upon a solution of sulphuretted hydrogen mixed 
 with starch paste ; or in the cessation of the formation of a 
 precipitate upon further addition of the standard solution, 
 as is the case when a standard solution of chloride of sodium 
 is used to precipitate silver from its solution in nitric acid ; 
 or in incipient precipitation , as is the case when a standard 
 solution of silver is added to a solution of hydrocyanic 
 acid mixed with an alkali ; or in a change in the action of 
 the examined fluid upon a particular reagent, as is the case 
 when a solution of arsenite of soda is added, drop by drop, 
 to a solution of chloride of lime, until the mixture no longer 
 imparts a blue tint to paper moistened with iodide of 
 potassium and starch paste, &c.' 
 
 The only condition on which the volumetric system of 
 analysis can be carried on successfully is, that the greatest 
 
302 APPARATUS FOR VOLUMETRIC ANALYSIS. 
 
 care is exercised with respect to the graduation of the 
 measuring instruments, and the strength and purity of the 
 standard solutions. A very slight error in the analytical 
 process becomes considerably magnified when calculated 
 for pounds, hundredweights, or tons of the substance 
 tested. The end of the operation in this method of analysis 
 is in all cases made apparent to the eye. (Button.) 
 
 STANDARD SOLUTIONS. 
 
 A very useful form of bottle for their preservation is 
 the ordinary wash-bottle, or any common bottle fitted with 
 the same arrangement of tubes. The mouth end of the 
 blowing tube should be furnished with a tightly fitting- 
 india-rubber cap to prevent alteration of the standard by 
 evaporation. A similar cap over the point will be useful, 
 although not absolutely necessary, or it may be closed by 
 a small cork fitting over it. Burettes can then be filled 
 with the solution without its frothing, and if the tube 
 which enters the liquid does not reach the bottom of the 
 bottle, the sediment, if any, is not disturbed ; another ad- 
 vantage is that the solution does not come into contact 
 with the cork, nor can any dust enter. 
 
 The Instruments and Apparatus. 
 
 The Burette, or graduated tube for delivering the stan- 
 dard solution, may be obtained in a great many forms 
 under the names of their respective inventors, such as Mohr, 
 Gay-Lussac, Binks, &c. ; but as some of these possess a de- 
 cided superiority over others, it is not quite a matter of in- 
 difference which is used, and therefore a slight description 
 of them may not be out of place here. The burette, with 
 india-rubber tube and clip, contrived by Dr. Frederic Mohr 
 of Coblentz, shown in figs. 82 and 83, has the preference 
 over all others for general purposes. 
 
 The advantages possessed by this instrument are, that 
 its constant upright position enables the operator at once 
 
APPARATUS FOR VOLUMETRIC ANALYSIS. 
 
 303 
 
 to read off the number of degrees of test solution used for 
 any analysis. The quantity of fluid to be delivered can 
 be regulated to the greatest nicety by the pressure of the 
 thumb and finger on the spring clip* and the instrument 
 
 FIG. 82. 
 
 FIG. 83. 
 
 not being held in the hand, there is no chance of increas- 
 ing the bulk of the fluid by the heat of the body, and thus 
 leading to incorrect measurement, as is the case with Binks's 
 or Gay-Lussac's form of instrument. The principal disad- 
 vantage of these two latter forms of burette is, that a correct 
 reading can only be obtained by placing them in an up- 
 right position, and allowing the fluid to find its perfect level. 
 
304 
 
 THE BURETTE. 
 
 FIG. 84. 
 
 FIG. 85. 
 
 The preference should, therefore, unhesitatingly be given to 
 Dr. Mohr's burette wherever it can be used ; the greatest 
 drawback to it is that it cannot be used for potassium 
 permanganate, in consequence of its india-rubber tube, 
 which decomposes the solution. 
 
 We are again indebted to Dr. Mohr for another form 
 of instrument to overcome this difficulty, viz. the foot 
 
 burette, with india-rubber 
 ball, shown in fig. 84. 
 
 The flow of liquid from 
 the exit tube can be regu- 
 lated to a great nicety by 
 pressure upon the elastic 
 ball, which is of the ordi- 
 nary kind sold for children, 
 and has two openings, one 
 cemented to the tube with 
 shellac, and the other at the 
 side, over which the thumb 
 is placed when pressed, and 
 on the removal of which it 
 refills itself with air. 
 
 Gay-Lussac's burette, 
 supported in a wooden 
 foot, may be used instead 
 of the above form, by in- 
 serting a good fitting cork 
 into the open end, through 
 
 which a small tube bent at right angles is passed. If the 
 burette is held in the right hand, slightly inclined towards 
 the beaker or flask into which the fluid is to be measured, 
 and the mouth applied to the tube, any portion of the solu- 
 tion may be emptied out by the pressure of the breath, and 
 the disadvantage of holding the instrument in a horizontal 
 position, to the great danger of spilling the contents, is 
 avoided ; at the same time the beaker or flask can be held 
 in the left hand and shaken so as to mix the fluids, and by 
 this means the end of the operation is more accurately 
 determined . 
 
THE BUEETTE. 
 
 305 
 
 FIG. 86. FIG. 87. 
 
 Fig. 85 will show the arrangement here described. 
 Mr. J. Blodget Britton has described, in the ' Chemical 
 
 News ' for August 5, 1870, a burette for use in determining 
 
 iron in metals and ores. 
 
 Figs. 86 and .87 represent two of the kind, but of 
 
 different patterns, mounted on walnut-wood stands; the 
 
 former is for metals, and 
 
 the latter, of smaller capa- 
 city, for ores. 
 
 Securely fastened to the 
 
 upright B is a graduated 
 
 tube A, having its lower 
 
 part drawn out in the 
 
 usual manner, but bent 
 
 outwards at an angle of 
 
 about 25. E is a piece of 
 
 cork riveted into a sheet 
 
 steel spring D, which presses 
 
 tightly, by means of the 
 
 latter, against the vent of 
 
 the tube, c is a thumb- 
 screw passing through the 
 frame and bearing against 
 the spring. 
 
 The tube of fig. 87 has 
 a capacity of 100 c.c., and 
 is graduated into tenths, or 
 1000 ; but that of fig.. 86 
 has a capacity of 150 c.c., 
 and is graduated into twen- 
 tieths, though only at its lower and narrow part. 
 
 Modes of operating. Place a small narrow-necked 
 funnel in the tube, as shown by the figures ; pour in the 
 solution to be used until it quite reaches the funnel, and 
 then remove the latter to carry away any floating bubbles ; 
 turn the thumbscrew and bring the top line of the solution 
 exactly to the zero line of the scale ; stop the flow, and after- 
 wards touch the point of the cork with a glass rod to take 
 from it any adhering drop. The instrument is then ready 
 
 x 
 
306 
 
 THE PIPETTE. 
 
 FIG. 88. 
 
 50CC 
 
 10 CC 
 
 for use. By means of the thumbscrew the dropping may be 
 controlled with extreme nicety or instantly stopped. Cork, 
 after a little use, becomes quite inert towards potassium 
 permanganate. The occasional application of some pure 
 tallow to the end of the tube and cork will be quite effectual 
 
 in preventing any of the fluid from 
 running upwards by capillary at- 
 traction. 
 
 For everyday use in the labo- 
 ratory, as well as for very accurate 
 determinations, this burette will 
 find favour. 
 
 The Pipette. The pipettes used 
 in volumetric analysis are of two 
 kinds, viz, those which deliver 
 one certain quantity ,only, and 
 those which are graduated so as to 
 deliver various quantities at the 
 discretion of the analyst. In the 
 former kind, or whole pipette, the 
 graduation may be of three kinds ; 
 namely, 1st, in which the fluid is 
 suffered to run out by its own 
 momentum only. 2nd, in which 
 it is blown out by the breath. 
 3rd, in which it is allowed to run 
 out to a definite mark. Of these 
 methods the last is preferable in 
 point of accuracy, and should 
 therefore be adopted if possible. 
 The next best form is that in 
 which the liquid flows out by its 
 own momentum, but in this case 
 
 the last few drops empty themselves very slowly ; but if 
 the lower end of the pipette be touched against the beaker 
 or other vessel into which the fluid is poured, the flow is 
 hastened considerably, and in graduating the pipette it is 
 preferable to do it on this plan. 
 
 In both the whole and graduated pipettes the upper 
 
COLOEIMETRIC ANALYSIS. 307 
 
 end is narrowed to about J inch, so that the pressure of 
 the moistened finger is sufficient to arrest the flow at any 
 point. 
 
 Fig. 88 shows two whole pipettes, one of small and the 
 other of large capacity, and also a graduated pipette of 
 medium size. 
 
 The Measuring Flasks. These indispensable instru- 
 ments are made of various capacities ; they serve to mix 
 up standard solutions to a given volume, and also for the 
 subdivision of the substance to be tested by means of the 
 pipettes, and are in many ways most convenient. They 
 should be tolerably wide at the mouth, and have a well - 
 ground glass stopper, and the graduation line should fall 
 just below the middle of the neck, so as to allow room for 
 vshaking up the fluid. 
 
 Colorimetric Analysis is also used in assaying. It is 
 based upon the fact that a coloured solution appears the 
 more intense the more of the colouring substance it con- 
 tains. 
 
 If, therefore, a solution containing a certain amount of 
 a substance, and being in consequence of a certain intensity 
 of colour, is prepared, it will be possible to obtain the solu- 
 tion under assay of an equal intensity of colour by appro- 
 priate dilution. 
 
 By measuring the volume of the assay solution and 
 taking into consideration the amount of the standard solu- 
 tion, the quantity of the substance contained in the assay 
 solution may be readily calculated. 
 
308 
 
 CHAPTEE IX. 
 
 THE ASSAY OF IRON. 
 
 THE ores of iron, properly so called, always contain the 
 metal in the oxidised state, and in various degrees of 
 purity. 
 
 The oxides and carbonates are the only minerals of 
 iron which can be used as ores in the blast-furnace. These 
 are associated with different impurities or foreign materials 
 in greater or less proportion. The following is a list of the 
 principal ores of iron, with the maximum percentage of 
 metallic iron that could occur in each if it were absolutely 
 pure, as in its formula : 
 
 MAGNETIC IRON ORE, Fe 3 4 =Fe 2 3 + FeO, 72-41. 
 
 RED HEMATITE and specular ore (anhydrous iron, ses- 
 quioxide or ferric oxide), Fe 2 3 , 70-00. 
 
 BROWN HEMATITE, limonite (brown iron ore), 2Fe 2 3 ,3H 2 0, 
 59-92. 
 
 SPATHIC IRON ORE (iron carbonate), FeO,C0 2 , 48-22. 
 
 TITANIFEROUS ORE (ilmenite), FeO,Ti0 2 , + nFe 2 3 . 
 
 FRANKLINITE, 3(FeO,ZnO,MnO) + (Fe 2 3 ,Mn 2 3 ), 45-16. 
 
 Besides these may be mentioned the varieties of impure 
 iron carbonates, known as clay or clay-band ironstone and 
 black-band ironstone. Clay-band ironstone sometimes re- 
 sembles compact limestone, sometimes greyish hardened 
 clay. Its great specific gravity, its effervescing on the addi- 
 tion of an acid, and acquiring a brown-red colour on roast- 
 ing, are sufficient means of identifying it. 
 
 The following is the result of an analysis of this class 
 of ore by the author ; the specimen was from Ireland, 
 county Leitrim : 
 
THE ASSAY OF IRON. 309 
 
 Ferrous oxide . 51-653 
 
 Ferric oxide 3'742 
 
 Manganese oxide -976 
 
 Alumina ....... 1-849 
 
 Magnesia -284 
 
 Lime -410 
 
 Potash -274 
 
 Soda -372 
 
 Sulphur . -214 
 
 Phosphoric acid -284 
 
 Carbonic acid . 31-142 
 
 Silica 6-640 
 
 Carbonaceous matter and loss . . . 2-160 
 
 100-000 
 
 Black-band is a combustible schistose variety of this ore. 
 The following analysis is also by the author : 
 
 Ferrous oxide ....... 20-924 
 
 Ferric oxide ....... -741 
 
 Manganese oxide . . . . . . 1-742 
 
 Alumina ........ 14*974 
 
 Magnesia . . . . . . -987 
 
 Lime . . . . . . . -881 
 
 Phosphoric acid ...... '114 
 
 Silica ..... "'"-. . . 26-179 
 
 Sulphur ........ -098 
 
 Carbonic acid . ..... . 14-000 
 
 Carbonaceous matter ..... 16*940 
 
 Water and loss 2-420 
 
 100-000 
 
 Besides these iron ores the following substances, con- 
 taining iron and used as fluxes, require assaying : Granite, 
 Chlorite, Basalt, Pyroxene; Amphibole, and also some kind 
 of slags (finery cinder, tap cinder, &c.) 
 
 A. THE ASSAY OF IRON IN THE DRY WAY. 
 
 Iron ores very seldom occur in a pure state, and the 
 ores may be arranged for their assay in the dry way (and 
 also for smelting) into five classes. 
 
 1. Iron ores containing silica, lime, and another base, 
 
 which ores are fusible per se. 
 
 2. Iron ores containing predominantly silica. 
 
 3. lime. 
 
 3. alumina. 
 
310 THE ASSAY OF IRON IN THE DRY WAY. 
 
 5. Iron ores containing a large amount of magnesia ;. 
 these ores are most difficultly rendered fluid. 
 
 The flux used for assaying (and also melting) varies 
 according to the nature of the predominant compound, and 
 the quantity used according to the amount of that com- 
 pound. 
 
 If the composition of the ore is known, it is easy to 
 ascertain the amount of flux necessary to form a slag with 
 the bases or silica present ; in most cases an extra quantity 
 of the flux should be added, in order to produce a sufficient 
 volume of slag to cover the button. 
 
 According to Dr. Percy,* blast-furnace cinder of the 
 following formula may be taken as a type of the kind of 
 slag desirable : 
 
 Al 2 3 ,Si0 3 2(3CaO,Si0 3 ). 
 Its approximate composition per cent, is as under : 
 
 Silica ..... 38"| T2 parts 
 
 Alumina . . . 15 > or about < 1 part 
 Lime .... 47; L3 parts 
 
 The following mixtures of various fluxes, when fused,, 
 produce a slag which may be regarded as approximating 
 to the above composition : 
 
 Quartz 
 
 
 
 1 
 
 1 
 
 ^| 
 
 T36-5 per cent. 
 
 China Clay 
 
 ". 
 
 2 
 
 f Silica . 
 \Alumina 
 
 0-92 / 
 0-82 
 
 0-82 f 
 
 J 
 = < 15*5 
 
 Lime . 
 
 . 
 
 2 
 
 . 
 
 . 
 
 2-5J 
 
 148-0 ' 
 
 Glass . 
 
 
 
 2* 
 
 I Silica . 
 ( Materials = Alumina 
 
 1-75^ 
 tO-75 \- 
 
 r35 
 = 4 16 
 
 Lime . 
 
 , 
 
 2^ 
 
 
 
 2-5 J 
 
 L50 
 
 
 
 Shale or fireclay 
 Lime . 
 
 3 
 
 f Silica . 
 \Alumina . . 
 
 wn 
 
 0-9 ^> 
 2-5 J 
 
 f35 
 - J 17 
 
 Us 
 
 > 
 
 According to Bodemann a compound of 56 % silica, 
 30 % lime, and 14 % alumina forms a slag most easily ren- 
 dered fluid, but as it is found that this slag itself is not 
 sufficiently fusible in a small assaying furnace (air-furnace), 
 an addition of fluor-spar is made to the mixture, and in 
 some cases (the iron ore being very difficultly rendered 
 
 * Percy's ' Metallurgy,' p. 240. 
 
 t 30 % say of alkalies, liine, &c., on account of their fusibility, are taken as? 
 equivalent to so much alumina. 
 
THE ASSAY OF IKO^ IN THE DEY WAY. 311 
 
 fluid) some borax is added, or a mixture of borax and fluor- 
 spar. 
 
 An exact knowledge of the mineralogical properties 
 of the iron ores, and a due experience, will enable the 
 assay er to properly adjust the fluxes without resorting to 
 an analysis to find what amount of silica and bases are 
 present. 
 
 In some iron-works of Germany the following propor- 
 tions of fluxes are used : 
 
 For magnetic ore, red hematite ~) r on n/ t, n r /a 
 
 (very rich) ' ' c " a ^ anc * 25 %fluor-spar 
 
 argillaceous brown iron ore 20 to 40 30 to 40 
 
 bog iron ore ... 50 50 
 
 spathose iron ore . . 10 to 15 20 to 25 
 
 finery cinder . . . 20 to 25 20 to 25 
 
 Air-furnaces are best adapted for assaying iron ores 
 where many assays are required. 
 
 The furnace should have a cross section of 18 in. x 18 
 in., and depth to grate-bars 21 inches. Flue 7 in. x 7 in. 
 Anthracite is the best fuel to use ; then coke. 
 
 In the assay of an iron ore it is required to reduce the 
 oxide of iron to the metallic state, or rather to that of cast 
 iron, to collect it in a button, and to form with the foreign 
 materials of the ore by means of fluxes a fusible slag; 
 that will not retain any of the iron in combination or in 
 the form of pellets. ; 
 
 Naked crucibles, either* of clay or black-lead, or cru- 
 cibles lined with charcoal, are employed ; the latter are 
 preferable. The button of metal does ,not adhere to naked 
 pots, but the slag adheres very strongly ; so much so that 
 it cannot be detached with any degree. of accuracy for 
 weighing (which in some of M. Berthier's processes is of 
 importance). Black-lead pots allow neither the slag nor 
 button to adhere, but the former dissolves much argilla- 
 ceous matter from the pot, so that its weight is greatly" 
 increased, and the assay cannot be verified. In naked 
 crucibles charcoal must always be added to the assay, to 
 reduce the iron oxide ; in which case, if an excess be.added, 
 it prevents the button from completely forming, so that 
 
312 THE ASSAY OF IRON IN THE DRY WAY. 
 
 globules remain in the slag (with care this may, however, 
 be avoided). Neither do naked crucibles resist the fire as 
 well as those lined with charcoal, because the lining sup- 
 ports the sides when they soften. The charcoal lining also 
 allows the assay to be finished without adding any reagent 
 to the ore ; the button can be readily taken out, because 
 it does not adhere to the charcoal ; and lastly, the earthy 
 matters in the ore, which have formed a slag, may be col- 
 lected and weighed. If we have added any flux to the ore, 
 the total weight can also be ascertained. 
 
 The method of lining crucibles with charcoal brasque 
 and conducting the assay is given as follows by Mr. 
 Blossom : * 
 
 The brasque has a composition of four parts of finely 
 pulverised charcoal to one part of molasses. This must 
 be thoroughly kneaded until a ball of it made in the hands 
 resists, to a sensible degree, an attempt to pull it apart. 
 The crucibles are packed full by driving the brasque in 
 with a mallet ; a conical cavity, of sufficient size for the 
 charge, is then cut out, and the brasque dried in an oven. 
 Care must be taken not to burn the molasses, for the 
 brasque would in that case crumble, and be useless. 
 
 These are the best crucibles for iron assays, because 
 they combine the following advantages : 
 
 Being lined with charcoal, none need be mixed with 
 the charge to reduce the oxide, whereas in naked crucibles 
 charcoal must be added, and is liable to prevent the com- 
 plete collection of the iron in a button by holding little 
 pellets in suspension. 
 
 The slag neither adheres to a charcoal lining nor takes 
 up any material from it, while it does adhere to ordinary 
 naked crucibles, and dissolves argillaceous matter from 
 black-lead crucibles. In the former case, then, the slag 
 may be weighed as a verification of the assay, while in the 
 latter this is, of course, impossible. 
 
 The lining serves as a support for the crucible, which, 
 under the high heat employed, is very apt to be softened 
 and crushed beneath the weight of the fuel. 
 
 * ' Chemical News,' April 5, 1872. 
 
THE ASSAY OF IRON IN THE DRY WAY. 313 
 
 The Charge. 
 
 In making up the charge it is only necessary to con- 
 sider the materials required as fluxes for the foreign 
 matters of the ores. It may be well to sprinkle a little 
 charcoal into the charge as a precaution, but none is abso- 
 lutely required. Two cases may arise, in which we have 
 (1) ores of unknown composition, and (2) ores previously 
 analysed. The assay in both cases gives us a clue to the 
 nature of the iron that may be obtained from the ore, and 
 to the character and proportion of the fluxes to be added in 
 the blast-furnace, in order that we may produce a fusible 
 slag free from iron. In the former case we obtain the 
 additional information of the approximate percentage of 
 iron, though the iron assay is seldom, if ever, made for 
 this purpose. Eecourse is had to the more accurate 
 chemical analysis, which gives us the exact proportions of 
 the substances which affect the iron injuriously or other- 
 wise. In all the assays a constant weight of ore, 300 
 grains, is taken. 
 
 1. Ores of Unknown Composition. 
 
 In the assay of an ore the composition of which is 
 unknown, we employ one or more preliminary assays, 
 and, if satisfactory results be not obtained from either, 
 we make another assay with a charge modified according 
 to the indications of the preliminary assay. The follow- 
 ing charges may be used to advantage in the preliminary 
 assay : 
 
 Preliminary Assay Charges. 
 
 i. Ii. m. iv. 
 
 Silica .... 75 30 120 75 grains. 
 
 Lime .... 75 120 45 75 
 
 Ore . . . . 300 300 300 300 
 
 1. The first charge is employed for the purer ores, those 
 containing very little gangue such as some varieties of 
 magnetic ore, red and brown hematites, specular and 
 micaceous ores. 
 
314 THE ASSAY OF IRON IN THE DRY WAY. 
 
 2. Ores containing silica ; some varieties of brown 
 hematite, magnetic ore, &c. 
 
 3. Ores containing lime, magnesia, or protoxide of man- 
 ganese carbonates, &c. ; calcareous hematites, spathic iron. 
 
 4. Ores containing silica and alumina ; clay ironstones, 
 black-band, &c. 
 
 The principle involved in all the charges is that of 
 furnishing for a base, lime ; for an acid, silica ; and vice 
 versa. 
 
 The choice of a charge, therefore, depends on the 
 acid or basic nature of the gangue of the ore. The mate- 
 rials of the gangue might possibly be associated in such 
 proportions as to flux themselves, but this would happen 
 rarely. 
 
 Ores containing titanium require the addition of fluor- 
 spar to the charge, in quantity varying from 15 to 300 
 grains, according to the amount of titanium present. 
 
 2. Ores previously Analysed. 
 
 When we know the percentage composition of an ore r 
 it is a very simple matter to calculate a charge for the dry 
 assay. Good results are obtained from a charge so propor- 
 tioned as to yield a slag corresponding to the following 
 formula of a blast-furnace cinder, as given by Percy : 
 
 H 2 3 represents alumina, and BD, lime, magnesia, and other 
 bases. Its approximate percentage composition is as fol- 
 lows : 
 
 Silica ...... 38 "| r 2^ parts 
 
 E 2 O 3 (alumina) . . . . 15 ^> or about < 1 part 
 
 KO (lime, magnesia, &c.) . . . 47 J [9 parts 
 
 We have, then, only to establish the latter relation 
 between the component materials of the gangue, to obtain,. 
 on fusion, the above slag. 
 
 Let us take the following example : 
 
 300 grains Ore Difference 
 
 The Ore contains Per cent. contain Required to be added 
 
 Silica .:....' 165 - 4'95 25' 20-05 grains. 
 
 Alumina . . 1-94 5-82 10- 4-18 
 
 Lime, MgO, &c. . 4'51 13-53 30- 6-47 
 
THE ASSAY OF IKON IN THE DRY WAY. 315 
 
 Silica is supplied by ground quartz. For the bases 
 EO represented in the furnace slag and in the ore by lime, 
 magnesia, &c., we add pure unslaked lime. The alumina 
 is added in the form of kaolin, which may be assumed to 
 contain equal parts of alumina and silica. Allowance must 
 be made in adding silica for that introduced with the 
 kaolin. 
 
 It happens sometimes that the ore contains more than 
 is required of one of the ingredients of the slags, or the 
 silica introduced with the kaolin may, when added to that 
 already present, increase the quantity beyond the re- 
 quirement. In either case make up a new slag with the 
 excess. 
 
 The charge having been weighed out, must be tho- 
 roughly mixed on glazed paper ; after placing it in the 
 crucible, the conical cavity is closed with a piece of char- 
 coal, and the whole top of the crucible is covered with a 
 luting of fireclay. The latter is mixed with ^ ^ part 
 fine sand, and is made plastic with borax water. Hair is 
 sometimes employed to prevent the luting from cracking 
 off when dry ; but no trouble is experienced from this 
 source if the luting be properly made and applied. It 
 should not be put on too thick, should be lapped over the 
 edges of the crucible, and thoroughly dried before placing- 
 the crucible in the furnace. .. 
 
 Four crucibles are introduced at one time, and rest 
 upon two fire-bricks placed one upon the other, to keep the 
 crucibles in the very midst of the glowing coals. If the 
 crucibles do not rest steadily on the bricks, it is well to 
 support them with a little luting, to prevent their being 
 knocked over in the fire. A low fire may be kindled before, 
 the introduction of the crucibles, or it may be kindled 
 around them. The fuel is added gradually until it fills 
 the furnace above the tops of the crucibles ; the fire is 
 then maintained at its maximum temperature for 2-J 3-| 
 hours, according to the refractory nature of the ore. Ores 
 containing much titanium may even require 4 hours, 
 while carbonates containing manganese may fuse well in 2-| 
 hours, or in less time. Three hours will generally be suifi- 
 
316 THE ASSAY OF IRON IN THE DRY WAY. 
 
 cient for ores that do not contain much titanium. When 
 the fire has burned out, the bricks and crucibles are 
 removed in one mass, cemented together by the slag of 
 the fuel. The crucibles are detached, and the exteriors 
 broken with a hammer ; on inverting and tapping the 
 brasque lining, the slag and the button of cast iron will 
 fall into the hand, when, if they adhere together, a slight 
 tap will suffice to separate them. Before separation, how- 
 ever, they ' should be carefully cleansed and weighed ; if 
 necessary, the slag may then be broken, and any particles 
 of iron it retains mechanically may be extracted with a 
 magnet. The weight of the iron being deducted from the 
 weight of the slag and button, we obtain the weight of the 
 slag. This ought to approximate closely to the weight of 
 the fluxes introduced and the corresponding material of 
 the ore. If a large amount of iron has combined with the 
 slag it will be indicated by the excess in weight. Titanium 
 and manganese enter the slag almost completely ; hence 
 if they are present, allowance must be made for them. 
 Duplicate assays are made, and the two results should 
 not differ more than 0'3 0-4 of one per cent. The slag 
 ought to be well fused, colourless, transparent, and vitreous, 
 or white, light-grey, bluish-grey , opaque, and semi-vitreous, 
 resembling porcelain or enamel. 
 
 A good button will be well formed, and will separate 
 completely from the slag. 
 
 If the metal be of good quality, the button, when. 
 wrapped in a piece of thin tin-plate, and struck on the 
 anvil, will flatten slightly before breaking. It ought to be 
 grey or greyish-white, and the grain fine, or tolerably fine. 
 A button of bad iron breaks readily without changing 
 form, sometimes even pulverising : the metal is generally 
 white and crystalline on the broken surface. 
 
 The following are some of the characters that may be 
 observed in slags, and their indications with reference to 
 the charge : A perfectly transparent slag of greenish tint 
 indicates an excess of silica. A stony rough slag, or one 
 that is crystalline in fracture and dull in lustre, indicates 
 an excess of bases. 
 
ASSAY OF IRON IN THE DRY WAY. 317 
 
 If the product, instead of being melted, is only fritted, 
 and contains the reduced iron interspersed as a fine grey 
 powder, both silica and alumina are deficient in the flux, 
 lime and magnesia being in excess. The latter is one of 
 the most refractory substances found in iron ores, and, 
 when present in quantity, requires an addition of both 
 silica and lime. A vesicular slag, with the iron interspersed 
 in malleable scales, indicates the presence in the ore, of iron 
 and manganese silicates, or an excess of silica, which react 
 on the carburetted iron as it forms, producing malleable 
 iron and carbonic acid. This trouble is corrected by the 
 addition of lime. 
 
 Manganese in small quantity gives an amethystine tint 
 to the slag ; in larger proportion it produces a yellow, 
 green, or brown colour. 
 
 Titanium often produces a resinous, black, and scoria- 
 ceous slag, sometimes curiously wrinkled on the outside. 
 It is covered on the outside with a metallic pellicle of the 
 cyano-nitride of titanium with its characteristic copper 
 colour ; sometimes the slag is vitreous and of a bluish tint. 
 Chromium gives a dark resinous slag, sometimes surrounded 
 by a thin metallic coating. 
 
 The following are some characters of the button depend- 
 ent on the substances named : 
 
 Phosphorus. A hard, brittle, white metal, known as 
 cold-short. 
 
 Sulphur. A strong, reticulated, mottled structure, and 
 red-short iron. 
 
 Manganese. A button smooth exteriorly, hard and 
 non-graphitic : it breaks under the hammer, and presents a 
 white crystalline fracture. 
 
 Titanium. The button is smooth on the outside, and 
 breaks under the hammer with a deep grey fracture, dull 
 or crystalline. It adheres strongly to the slag. The button 
 is covered sometimes with titanium cyano-nitride with its 
 characteristic copper colour. Titanium is said to increase 
 the strength of the metal. It may be present to the extent 
 of one per cent. 
 
 Chromium. Sometimes the button is smooth, well 
 
318 THE ASSAY OF IRON IN THE DRY WAY. 
 
 fused, with a brilliant crystalline fracture and tin- white 
 colour ; at other times it is white, only half fused, or it 
 may even form a spongy mass of a clear grey colour, 
 according to the quantity of chromium contained in the 
 iron. Many alloys of iron and chromium will scratch 
 glass. 
 
 Berthier recommends the following method for esti- 
 mating the other, chiefly slag-forming, components of iron 
 ores. The operations of this method are comprised in 
 roasting or calcining, to drive off any volatile or combust- 
 ible matters, and in treating the ore with certain acids, 
 the object of which is to ascertain the amount of insoluble 
 matter, by difference of weight, before and after the action 
 has taken place. 
 
 The hydrated ores are calcined to estimate water ; 
 and those containing manganese, to reduce it to a fixed and 
 known state of oxidation (sesquioxide). The carbonates 
 are roasted to expel carbonic acid, and the ores from the 
 coal formations to burn the combustible matter with which 
 they are mixed. 
 
 Slags and dross are also roasted to free them from 
 charcoal. A simple calcination sometimes is sufficient, as in 
 the case of carbonates ; but where mixtures of ferric and 
 ferrous oxides are to be assayed, they must be subjected 
 to a long roasting, in order to convert all the contained 
 ferrous oxide into ferric oxide. 
 
 Diluted and cold nitric or acetic acid is employed 
 for minerals whose matrix is purely calcareous or magne- 
 sian, as these acids dissolve the earthy carbonates, without 
 attacking either stones, clay, or the iron oxides. The 
 residue is to be well washed, dried, and weighed, and the 
 amount of carbonates calculated by the difference. It is 
 now to be treated with boiling hydrochloric acid, or, what 
 is preferable, aqua regia. The ores which contain sub- 
 stances insoluble in these acids are generally of a clayey or 
 flinty nature. These are to be weighed, and according to 
 their weight that of the flux to be added in the assay is 
 estimated, as will be shown hereafter. 
 
 It must be borne in mind, however, that the clays 
 
THE ASSAY OF IRON IN THE DRY WAY. 319 
 
 are not absolutely insoluble in hydrochloric acid, for a 
 certain quantity of alumina is always dissolved, which is 
 generally greater in proportion to the amount present in 
 the iron ore. 
 
 The ores containing titanium are boiled with concen- 
 trated sulphuric acid, after they have been reduced to the 
 finest possible state of division. All the iron, titanium, and 
 manganese oxides are dissolved, and the stony gangues 
 which resist the action of this acid can be estimated. 
 The utility of this estimation will be pointed out as we 
 proceed. 
 
 When all the operations necessary for each particular 
 case have been completed, we know the proportion of vola- 
 tile substances, of substances soluble in acetic acid, and 
 those insoluble in hydrochloric and sulphuric acids, con- 
 tained in the substance under assay. 
 
 Let A be the weight of the rough or non-calcined ore ; 
 B the weight of the same calcined ; C the weight of the 
 fluxes in a rough state ; D the weight of the same calcined ; 
 P the weight of matter insoluble in hydrochloric or sul- 
 phuric acid ; E the weight of the fixed substances soluble 
 in acetic or nitric acid a weight which can be readily 
 calculated when we know the loss which the ore, not 
 treated by acids, suffers by calcination, and the residue of 
 the treatment of this substance by acetic or nitric acid ; M 
 the weight of the button of metal and scattered globules ; 
 S the weight of the slag ; and the loss of weight in the 
 assay, which represents the quantity of oxygen disengaged 
 during the reduction. 
 
 The following is the disposition of the data from which, 
 at one view, all the useful results of the assay can be de- 
 termined. 
 
 In the assay has been employed 
 
 A, rough ore = calcined ore . . . . B 
 
 B, of rough fluxes added = fixed flux . . D 
 
 Total of fixed matter B + D 
 
 The result has been 
 
320 THE ASSAY OF IRON IX THE DRY WAY. 
 
 Metal M\ T , i M 
 
 Slag-S / T tal M + 
 
 Loss 
 
 Fluxes D 
 
 Verifiable matters S D 
 
 Substances insoluble in hydrochloric acid, &c. . T 
 Substances soluble in hydrochloric acid, &c. . S D T 
 
 Substances soluble in acetic acid ... R, 
 
 Substances insoluble in acetic acid, and solu- 
 ble in hydrochloric acid .... S D T R 
 
 When the iron in the substance assayed is in a known 
 degree of oxidation, and when but little manganese is 
 present, the quantity of oxygen ought to correspond 
 very nearly with the quantity of metal M produced : if it 
 does, the assay must be correct. 
 
 A rigorous correspondence between the two numbers, 
 however, cannot always be obtained, because the iron is 
 not pure, but always contains carbon, so that in ordinary 
 assays the ferric oxide loses but from twenty-eight to 
 twenty-nine per cent, of oxygen. 
 
 On the other hand, the quantity of iron remaining in 
 the slag makes up in part for the carbon combined with 
 the metal reduced ; but when the assay has been made 
 with a suitable flux, the quantity of oxide remaining is 
 very small, and never exceeds one per cent, of the weight 
 of the slag. When the iron is in an unknown degree 
 of oxidation, the loss produced in the assay gives the 
 degree, if it has been made without accident ; but if there 
 is any doubt, and the result is of importance, the assay 
 must be recommenced for verification. If the ferruginous 
 matter contain manganese, and if that metal be in the 
 state of protoxide, the verification just described can be 
 made without modification, because the manganese dis- 
 solved in the slag is always at the minimum of oxidation ; 
 and when a sufficient quantity of flux is employed, the 
 amount reduced is of no consequence. But when the 
 manganese is in the state of red oxide, it parts with a 
 certain quantity of oxygen on being reduced to the 
 minimum of oxidation f quantity estimated in the loss 
 0), so that a perfectly accurate verification cannot 
 be made. Nevertheless, the difference between the loss 
 
THE ASSAY OF IKON IN THE WET WAY. 821 
 
 0, and the quantity of oxygen calculated from the metal 
 M, cannot be very great, because the red oxide of man- 
 ganese loses but *068 of oxygen in its transformation to 
 protoxide. 
 
 If the assay has been made with care, the loss of 
 oxygen indicates the amount of iron in a very approxi- 
 mate manner, and nearly always with an exactitude 
 which is surprising to those not accustomed to this kind 
 of operation. 
 
 Titanic acid behaves in iron assays exactly as the 
 oxides of manganese ; it disengages at most but -06 of 
 oxygen when dissolved in the earthy glasses in contact 
 with charcoal. 
 
 B. THE ASSAY OF IRON AND ITS OEES IN THE WET WAY. 
 
 This will be subdivided into the following sections : 
 
 a. Assay of iron ores, pig iron, and steel, for the 
 metallic iron they contain. 
 
 $. Complete assay of iron ores. 
 
 7. Estimation of carbon, sulphur, phosphorus, silicon, 
 &c., or metallic iron and steel. 
 
 a. Dr. Penny's Process. The following method of 
 estimating the amount of iron in a sample by means of a 
 normal solution has been contrived by Dr. F. Penny, who 
 was led to substitute potassium bichromate for potassium 
 permanganate, as recommended by Marguerite. The 
 reason of employing the bichromate is that it is an 
 unchangeable salt, whilst the permanganate sometimes 
 undergoes decomposition, so that its strength is variable, 
 and each series of experiments made with it requires a 
 separate verification by means of a weighed quantity of 
 pure iron. This inconvenience is avoided in Dr. Penny's 
 method, which is described in his own words as under : 
 . ' I shall proceed to describe the method of employ- 
 ing the potassium bichromate for the determination of the 
 amount of iron in clay-band and black-band ironstone. I 
 shall be purposely minute, as I particularly desire that 
 the process may be serviceable to those who, from their 
 
 Y 
 
322 THE ASSAY OF IRON IN THE WET WAY. 
 
 pursuits in life, are interested in the value and quality of 
 ironstone, and who may be imperfectly acquainted with 
 analytical operations. 
 
 4 A convenient quantity of the specimen is reduced to 
 coarse powder, and one-half at least of this still further 
 pulverised, until it is no longer gritty between the fingers. 
 The test solution of potassium bichromate is next prepared. 
 44*4 grains of the salt in fine powder are weighed out, and 
 put into an alkalimeter (graduated into 100 equal divisions), 
 and tepid distilled water afterwards poured in until the 
 instrument is filled to 0. The palm of the hand is then 
 securely placed on the top, and the contents agitated by 
 repeatedly inverting the instrument, until the salt is dis- 
 solved and the solution rendered of uniform density 
 throughout. It is obvious that each division of the 
 solution thus prepared contains 0-444 grain of bichromate, 
 which corresponds to ^ a grain of metallic iron. The 
 potassium bichromate used for this process must of course 
 be purchased pure, or made so by repeated crystallisation, 
 and it should be thoroughly dried by being heated to 
 incipient fusion. 
 
 ' 100 grains of the pulverised ironstone are now intro- 
 duced into a Florence flask, with 1^ oz. by measure of 
 strong hydrochloric acid, and \ an ounce of distilled 
 water. Heat is cautiously applied, and the mixture occa- 
 sionally agitated, until the effervescence caused by the 
 escape of the carbonic acid ceases ; the heat is then 
 increased, and the mixture made to boil, and kept at 
 moderate ebullition for ten minutes or a quarter of an 
 hour. During these operations it will be advisable to 
 incline the flask, in order to avoid the projection, and 
 consequent loss, of any portion of the liquid by spirting. 
 About 6 oz. of water are next added, and mixed with the 
 contents of the flask, and the whole rapidly transferred to 
 an evaporating basin. The flask is rinsed several times 
 with water, to remove all adhering solution. 
 
 '.Several small portions of a weak solution of pure red 
 potassium prussiate (containing one part of the salt to 40 
 of water) are now dropped upon a white porcelain slab, 
 
DE. PENNY'S PROCESS. 323 
 
 which is conveniently placed for testing the solution in 
 the basin during the next operation. 
 
 ' The prepared solution of potassium bichromate in the 
 alkalimeter is then added very cautiously to the solution of 
 iron, which must be repeatedly stirred, and as soon as it 
 assumes a dark greenish shade it should be occasionally 
 tested with the red potassium prussiate. This may be 
 -easily done by taking out a small quantity on the top of a 
 glass rod, and mixing it with a drop of the solution on a 
 porcelain slab. When it is noticed that the last drop 
 communicates a distinct red tinge the operation is ter- 
 minated. The alkalimeter is allowed to drain for a few 
 minutes, and the number of divisions in the test-liquor 
 consumed read off. This number multiplied by two gives 
 the amount of iron per cent, in the specimen of ironstone, 
 assuming that, as directed, 100 grains have been used for 
 the experiment. The necessary calculation for ascertain- 
 ing the corresponding quantity of protoxide is obvious. 
 
 ' When black-band ironstone is the subject of ana- 
 lysis, or when the ore affords indication, by its appear- 
 ance or during the treatment with hydrochloric acid, that 
 it contains an appreciable quantity of carbonaceous 
 matter, then the hydrochloric acid solution must be 
 filtered before being transferred to the basin, and the 
 filter, with the insoluble ingredients, must be washed in 
 the usual way with warm distilled water, slightly acidu- 
 lated with hydrochloric acid until the filtrate ceases to 
 give a blue colour with the red potassium prussiate. 
 In those cases, also, where the presence of iron pyrites in 
 the ironstone is suspected, it will be necessary to remove 
 the insoluble matter by filtering before applying the 
 bichromate solution ; but with ironstones in which the 
 insoluble ingredients are merely clay and silica, filtration 
 is not essential. 
 
 6 Now it is evident" that the foregoing process, so far as 
 I have described it, serves for the determination of that 
 portion of iron only which exists in the ore in the state 
 of protoxide. But many specimens of the common iron- 
 stone of this country contain appreciable quantities of 
 
 T 2 
 
324 THE ASSAY OF IRON IN THE WET WAY. 
 
 ferric oxide, which, being unacted upon by the potas- 
 sium bichromate, would escape estimation by the present 
 method. By an additional and easy operation, however, 
 the amount of metallic iron in the ingredient may be 
 likewise determined. It is only necessary to reduce it 
 to the minimum state of oxidation and then to add the 
 bichromate as previously directed. 
 
 ' The best and most convenient agent for effecting the 
 reduction of the ferric oxide is sodium sulphite. The 
 only precaution to be observed is to use it in sufficient 
 quantity, and at the same time to take care that the iron 
 solution contains excess of acid. When the reduction is 
 complete, a few minutes' ebullition suffices to decompose 
 the excess of sodium sulphite, and effectually to expel 
 every trace of sulphurous acid. 
 
 ' In order to test the exactness of this mode of 
 estimating the iron of the peroxide, I made several ex- 
 periments with peroxide prepared from known quantities 
 of pure iron wire. The peroxide was thoroughly washed, 
 dissolved in hydrochloric acid, reduced with sulphite 
 of soda, and after complete expulsion of the excess of 
 sulphurous acid, the solution was diluted with water and 
 treated with potassium bichromate. I select three of the 
 experiments : 
 
 ' Exp. I. 10 grs. of iron consumed 8-87 of bichromate. 
 II. 18 15-94 
 
 III. 25 22-15 
 
 ; The mean of all my experiments on this point gives 
 the ratio of 100 of iron to 88'6 of bichromate. 
 
 ' Whenever, therefore, the ore of iron contains ferric 
 oxide it will be necessary to add sodium sulphite to the 
 hydrochloric acid solution before the addition of the test- 
 liquor from the alkalimeter. The sulphite should be dis- 
 solved in distilled water, and added to the solution of iron 
 in small successive portions, until a drop of the liquor 
 gives merely a rose-pink colour with potassium sulpho- 
 cyanide, which indicates that the reduction of the ferric 
 salt is sufficiently perfect. The liquor is now heated till 
 the odour of sulphurous acid is no longer perceptible. 
 
DR. PENNY S PROCESS. 325 
 
 These operations should be performed while the solution 
 is in the flask, and before it is filtered or transmitted to 
 the basin. 
 
 ' I will here mention, for the guidance of those who 
 may not be fully aware of the reactions of the oxides of 
 iron, that the existence of an appreciable quantity of 
 peroxide in the ironstone may be readily discovered by 
 dissolving (as directed in the process) 39 or 40 grs. of the 
 ore in hydrochloric acid, diluting with about 8 oz. of 
 water, filtering, and testing a portion of the solution with 
 potassium sulphocyanide. If a decided dark blood-red 
 colour is produced, the quantity of ferric oxide in the 
 stone must be determined ; but if the colour is only light 
 red or rose-pink, the proportion is exceedingly small, and 
 for practical purposes not worth estimating. Of course, 
 when the specimen of ironstone has an ochrey or a reddish 
 appearance on the surface or in the fracture, the presence 
 of a large proportion of ferric oxide is indicated, and its 
 exact quantity must be determined.' 
 
 The details of Dr. Penny's process have been carefully 
 examined by Mr. R. W. Atkinson, with the result of elimi- 
 nating several small sources of error which interfered 
 with accuracy. In all cases where volumetric methods 
 are employed, the first and most important point to be 
 considered is how to obtain the standard solution invari- 
 ably of an accurately known strength, and it is from a 
 variation in the methods employed for this end that a large 
 amount of the variation in the results is due. Presuming 
 that potassium bichromate is the salt almost universally 
 employed in England, it seems simple enough to weigh out 
 the exact amount of salt and dissolve it in a known volume 
 of pure distilled water. Dr. Penny recommends the fusion 
 of the bichromate for the purpose of driving off com- 
 pletely the water entangled in the crystals ; but if this be 
 done, however carefully the heat be regulated, on dis- 
 solving in water, and allowing to settle, a green deposit of 
 chromic oxide will be found at the bottom of the flask or 
 bottle, showing that a small amount of decomposition 
 takes place during fusion, and that the value of the stan- 
 
326 THE ASSAY OF IKON IN THE WET WAY. 
 
 dard must be lower than the calculated value. Instead of 
 fusing the salt, it is much better to grind up the very 
 purest crystals and to dry them in a steam oven for 
 several hours before weighing out. But, however the 
 solution be made up, it is always safer to standardise it 
 before use. 
 
 The great difficulty in the use of salts of iron for stan- 
 dardising is to obtain one which will not alter in composi- 
 tion by keeping for two or three months. Ferrous sulphate 
 is out of the question ; and most specimens of the iron and 
 ammonium sulphate are liable to oxidation, though in a 
 less degree than the ferrous sulphate. We have been for- 
 tunate enough to get from Messrs. Hopkin and Williams 
 a sample of granulated iron and ammonium sulphate, 
 which after several months' use still gives the theoretical 
 percentage of iron, ammonia, and sulphuric acid. Pre- 
 suming that the strength of the bichromate solution in 
 terms of iron is accurately known, a fair sample of the 
 ore having been ground sufficiently fine to pass com- 
 pletely through a sieve of 120 meshes to the linear inch, 
 part of it is dried at 212, and when cold portions of about 
 fifteen or sixteen grains are weighed out for analysis. 
 Unless the balance is very rapid in its action, it will be 
 found necessary to re-dry the portions first weighed out 
 if accurate and concordant results are to be expected.. 
 The ore is in such a very fine state of division that it 
 greedily absorbs moisture from the air during the opera- 
 tion of weighing out : this applies most forcibly to the 
 soft red ores, like Campanil and Vena Dulce, but it is also 
 true of the harder Eubio and other brown ores. The 
 weight of the portion being accurately known, it is trans- 
 ferred to a conical flask, and digested at a gentle heat on 
 a hot plate, with from 150 to 230 grains strong hydro- 
 chloric acid, the flask being closed with a watch-glass. 
 
 It has been asserted by K. F. Fohr that ferric chloride 
 is volatile at about 212, but this is contrary to experience. 
 It is true that yellow drops of ferric chloride may some- 
 times be seen depending from the under surface of the 
 watch-glass, but this is only the case when the solution is 
 
REDUCTION BY ZINC. 327 
 
 accompanied by a boiling of the liquid, and is doubtless 
 due to spirting. If the digestion is carried on so as to 
 avoid bubbling, the drops on the cover are always colour- 
 less, and no loss of iron from volatilisation need be feared. 
 
 When the ore has been completely dissolved, the next 
 step is the reduction from the ferric to the ferrous state, 
 to effect which there are three reducing agents mainly em- 
 ployed, viz. zinc, stannous chloride, and sulphurous acid 
 in the form of one of its salts. These will be dealt with in 
 the order given. 
 
 Eeduction by means of zinc is capable of giving 
 trustworthy results, provided that pure zinc free from 
 iron is employed ; or if the zinc contains iron, provided 
 that the amount thus introduced be allowed for. But the 
 use of zinc is open to other objections : in the first place, 
 it dissolves only slowly, and thus unduly retards the ope- 
 ration, which, when a number of analyses are to be carried 
 out, is a matter of no small moment. In the second place, 
 the titration is further retarded by the slowness with 
 which the blue colour with potassium ferricyanide is deve- 
 loped, in consequence of the presence of zinc chloride in 
 the solution. And lastly, the colour towards the end of 
 the titration becomes so faint, even when fully developed, 
 that it is impossible to distinguish the presence of an 
 amount of iron less than one or two tenths per cent, of 
 the iron contained in the ore. Consequently, although 
 reduction by means of zinc permits the analyst to obtain 
 uniformly concordant results, without the risk of error, its 
 action is too slow, and its indications are not sufficiently 
 delicate for the most accurate work. 
 
 But, however much rapidity of work may be an object 
 to be kept in view, there can be no doubt but that reduc- 
 tion by zinc is greatly to be preferred to the second 
 method, viz. reduction by means of stannous chloride. 
 Indeed, the latter method, as usually performed, is open 
 to the grossest abuses, and ought to be prohibited by any 
 authority which may seek to reform the present methods 
 of analysis. 
 
 One reason why stannous chloride has become a favour- 
 
3-28 THE ASSAY OF IRON IN THE WET WAY. 
 
 ite reducing agent is the rapidity with which the analysis 
 can be made. Starting with a sample of ore the percentage 
 of moisture and iron in the ore can be found by this 
 method (with the above qualifications as to accuracy) in 
 from two to three hours ; but it is contrary to the best 
 interests of a chemist to seek rapidity of work at the 
 expense of accuracy, and the abandonment of this method 
 of reduction is strongly recommended. 
 
 Of all methods, reduction of the ferric salt by the use 
 of a concentrated solution of ammonium bisulphate is the 
 most accurate and trustworthy. Sodium bisulphite is some- 
 times used, but is not nearly so satisfactory as the am- 
 monium salt, as it is more difficult to separate the last traces 
 of sulphurous acid from the former than from the latter. 
 The mode of manipulation is as follows : The ore having 
 been dissolved in hydrochloric acid in the conical vessel, 
 the solution is diluted with acidified water and filtered 
 into pear-shaped flasks, the filters being thoroughly washed 
 with hot acid water. The filtered ferric chloride is next 
 carefully neutralised with ammonia, strong at first and 
 afterwards dilute, until a faint reddish precipitate remains 
 permanent. Two or three drops of strong hydrochloric 
 acid are washed round the inner neck of the flask, and as 
 the acid flows down it spreads out, dissolving any par- 
 ticles of ferric hydrate which may have remained on the 
 sides of the flask. When the solution is quite clear, and 
 of a faint reddish colour, 75 to 90 grains of a strong solu- 
 tion of ammonium bisulphite (sp. gr. 1-06) are added, the 
 flask shaken, and boiling water added. On shaking the 
 flask the colour entirely disappears, and the flask is then 
 put over a burner. A small piece of thick platinum 
 wire is introduced to assist the boiling, and about 25 or 
 30 grains of dilute sulphuric acid (1 acid to 6 water) are 
 added to acidify the solution and to assist in the expulsion 
 of the excess of sulphurous acid. After the liquid is once 
 in a state of ebullition it is kept boiling briskly for thirty 
 minutes (less time is sufficient, but it is always well to err 
 on the safe side), during which time nearly the required 
 amount of bichromate is run out into the dish. At the 
 
REDUCTION BY SULPHUROUS ACID. .320 
 
 end of the half-hour the boiled solution is added to the 
 potassium bichromate, and the titration is carried out as 
 usual. 
 
 By proceeding as above there is only one loophole for 
 the introduction of error, viz. in the length of time allowed 
 for boiling off the sulphurous acid ; but if the conditions 
 as given above are fulfilled, constant and accurate results 
 may be relied upon. The above method has this advantage, 
 viz. that the solution is practically one of ammonio-ferrous 
 sulphate, a salt which is one of the most stable of all the 
 ferrous salts. It is, therefore, less liable to become oxi- 
 dised by exposure to the air in transferring to the basin 
 than the acid solution of ferrous chloride obtained by the 
 two previous modes of reduction. A further advantage 
 lies in the fact that the end reaction with potassium ferri- 
 cyanide is beautifully clear and delicate ; so that -there is no 
 difficulty in distinguishing the addition of ^ c.c. of bichro- 
 mate (strength 1 c.c. = O077 grain iron), equivalent to 0-0038 
 iron. Numerous experiments have shown that perfectly 
 constant results can be obtained by this method, the same 
 percentages of iron in a given ore having been found with 
 different standard bichromate solutions after the lapse of 
 several months. It is rarely the case that three experi- 
 ments carried out simultaneously give percentages of iron 
 differing by more than 0' 05 to 0-07 per cent, of the ore ; 
 but the main advantage which the method of reduction 
 by ammonium bisulphite possesses is that results are found 
 which are perfectly independent of any unconscious bias 
 on the part of the operator, and the author feels convinced 
 that were this method constantly and generally used, we 
 should hear less of differences of two per cent, and three 
 per cent, beween two chemists' analyses of the same sample 
 of ore. The existence of ' sellers ' and ' buyers ' chemists 
 is a disgrace to the profession, and anything which is likely 
 to put an end to the scandal, even in one trade, ought to 
 be welcomed. 
 
 One other point in the volumetric estimation of iron 
 remains to be noticed. The solution of potassium ferri- 
 cyanide slowly decomposes when the bottle containing it 
 
330 TITRATION OF IRON 
 
 is exposed to diffused daylight, and a yellowish sediment 
 is deposited. This change is very greatly retarded, if not 
 entirely prevented, by protecting the solution from light 
 by covering the bottle with an inverted tin canister. Con- 
 nected with this decomposition of the ferricyanide solution 
 is the fact, already well known, that when a mixture of 
 that solution with one of ferric chloride is exposed to day- 
 light reduction takes place, and the solution turns blue. 
 But it is not so generally known how rapidly this takes 
 place, and that if the drops of ferricyanide to which the 
 completely oxidised iron solution has been added be allowed 
 to remain exposed to the light (protected from dust by 
 means of a glass plate) for ten or fifteen minutes, a distinct 
 blue tinge will be developed. It is important to remember 
 this, for in titrating the blue colour requires two or three 
 minutes to become fully developed when the amount of 
 ferrous salt remaining in the solution is very small, and in 
 order to prevent the drop turning blue by reduction under 
 the influence of daylight, it is advisable to keep the slab 
 covered with a black cloth or with a flat tin cover. By 
 this means it is possible so to protect the mixture from 
 change that no blueness is perceptible after the lapse of 
 an hour or more, provided that all the iron in the solution 
 to be titrated has been fully oxidised. 
 
 TITRATION OF IRON WITH SODIUM HYPOSULPHITE. 
 
 A. C. Oudemans, jun., has proposed to estimate iron 
 in the acid solution of the chloride to which a little solu- 
 tion of copper sulphate and potassium sulpho-cyanide has 
 been added by dropping in a solution of sodium hyposul- 
 phite of known strength until the red colour of the iron 
 sulpho-cyanide has disappeared, and estimating the excess 
 of hyposulphite by titrating back with iodine solution. 
 
 A. E. Haswell modifies this method so as to avoid the 
 possibly disturbing separation of copper sulpho-cyanide, 
 and to dispense with the back titration with iodine. Ac- 
 cording to his experiments, the iodine-starch reaction often 
 takes place too early, before all the hyposulphite has been 
 converted into sodium tetrathionate. He explains this 
 
WITH SODIUM HYPOSULPHITE. 331 
 
 occurrence by the tendency of cupric iodide to split up 
 into cuprous iodide and free iodine, and thus produce a 
 premature blue coloration which after a time disappears 
 again as the cuprous iodide recornbines wdth the free iodine 
 to form the cupric iodide. 
 
 Haswell mixes the moderately acid solution of ferric 
 chloride in presence of a cupric salt with a few drops of 
 a dilute solution of sodium salicylate, and then reduces 
 with sodium hyposulphite. The deep violet colour of the 
 solution fades gradually and becomes colourless in presence 
 of a very slight excess of the reducing agent. The excess 
 of sodium hyposulphite is then oxidised with a dilute solu- 
 tion of sodium bichromate. The limit of the reduction is 
 sharply marked by the faint violet colour which indicates 
 the oxidation of a trace of the iron. It must be remem- 
 bered that strong hydrochloric acid destroys the colour 
 produced by salicylic acid in ferric chloride, which, how- 
 ever, is restored on moderate dilution with water. 
 
 For the execution of the method there are required a 
 solution of sodium hyposulphite, standardised by means of 
 a solution of ferric chloride of known strength ; a solution 
 of potassium bichromate about half the strength of the 
 sodium hyposulphite ; a solution of copper, prepared by 
 dissolving two grammes cupric ammonium chloride in 
 100 c.c. water ; and a solution of sodium salicylate con- 
 taining about 5 grammes of the salt per litre. 
 
 Five or ten c.c. of the iron solution are measured into 
 a small flask, slightly acidulated with hydrochloric acid, 
 and mixed with 1 to 2 c.c. of the copper solution, and a 
 few drops of the sodium salicylate. If the colour resulting 
 is not a pure violet, but an olive brown, the liquid is 
 diluted with water and the hyposulphite is added until the 
 liquid appears perfectly colourless on standing with the 
 back to the window and looking through the flask at a 
 sheet of white paper. It often happens that on adding 
 more sodium salicylate a faint coloration reappears, but 
 it is removed by a drop of .hyposulphite. It is then titrated 
 back with the bichromate until a faint viftlet coloration 
 appears. 
 
332 COMPLETE ASSAY OF IRON ORES. 
 
 Complete Assay of Iron Ores. The methods employed 
 in the analysis of iron ores have been thoroughly investi- 
 gated by Mr. A. A. Blair. The following account of the 
 method he recommends is taken from ' Mining Industries 
 of the United States.' The treatment of samples of iron 
 ore naturally divides itself into two parts, the mechanical 
 and the chemical, and it will be described under these 
 heads. The care with which the identity of every sample 
 is preserved throughout will be shown, and the methods by 
 which the estimation of the different elements is rendered 
 as accurate as the state of chemical knowledge would allow 
 will be given in detail. 
 
 THE MECHANICAL TREATMENT. 
 
 The sample is first taken from the bag and placed 
 upon a large piece of clean strong paper, and the label 
 removed from the box and put in a little card-rack fixed 
 over the bench. Half a dozen small chips, representing, 
 as nearly as possible, the different varieties of the ore, are 
 put aside and carefully labelled, to be used for specific 
 gravity estimations and for thin sections for the micro- 
 scope. 
 
 The large steel mortar, fig. 89, is half filled with the ore, 
 the leather cover adjusted, and the machinery started- 
 This mortar was cast at the Mid vale Steel Works of Phila- 
 delphia, and was made of an exceptionally fine quality of 
 propeller steel, containing over one per cent, of carbon and 
 about fifteen-thousandths of one per cent, of phosphorus. 
 The pestle was forged and hardened, and the wear, after 
 crushing between 15,000 and 20,000 pounds of ore, was 
 scarcely perceptible in either mortar or pestle. The mortar 
 weighs about 70 pounds, and the pestle with the stem and 
 weight about 25 pounds. The tappets A are faced with 
 raw hide, which stand the wear of the cams, H, remarkably 
 well, much better than either hard or soft iron or steel, 
 the dust from the ore causing the latter to cut very fast. 
 The shaft makes about 90 revolutions per minute, and the 
 ore, when the mortar is about half filled, feeds itself, so 
 
THE MECHANICAL TREATMENT. 
 
 333 
 
 that without any attention 25 pounds of hard ore in lumps 
 are reduced almost to powder in .about one and a half 
 hours. 
 
 The stem B is hooked up, so that the pestle clears the 
 
 FIG. 89. 
 
 top of the mortar A. The pulley D raises the mortar 
 clear of the block, and by means of the traveller E the 
 mortar is emptied on the chilled iron plate F, as figured 
 in the sketch. The ore is ground to a fine powder on this 
 chilled plate by the hardened steel muller c, any of it 
 falling off the plate being caught in the sheet-iron 
 troughs G. 
 
 It is thoroughly mixed and quartered, and the result- 
 ing sample, reduced finally to about 6 to 8 ounces, is 
 placed in a clean dry bottle. This bottle has the number 
 of the sample etched upon it, the same number in figures 
 half an inch high pasted on the neck, and the label which 
 came in the sample bag pasted on its side. The bottle 
 is not taken into the grinding-room until the sample is 
 being ground, and is always previously numbered, so 
 that when it is brought upstairs with the label the num- 
 bers can be compared and the label pasted on. A por- 
 
334 
 
 COMPLETE ASSAY OF IRON ORES. 
 
 tion of the sample is ground in the agate mortar A (fig. 
 90), with an agate pestle B, fitted in a flexible shaft c, and 
 
 FIG. 90. 
 
 CEILINC LINE 
 
 revolving at the rate of 700 times a minute, transferred to 
 a ground-glass stoppered bottle, and dried at 100 C. It 
 is then ready for analysis. 
 
 THE CHEMICAL TREATMENT. 
 
 Estimation of Phosphoric Acid. 
 
 For ores low in phosphoric acid 10 grms. are dissolved 
 in hydrochloric acid ; the solution is evaporated to dry- 
 ness on the sand bath, re-dissolved in dilute hydrochloric 
 acid (one part of acid to two of water), filtered, the filtrate 
 treated with 10 c.c. ammonium sulphite * to reduce 
 the iron to the ferrous condition, ammonia added until 
 the solution is nearly neutral, and then heated until it has 
 decolourised. Five c.c. of strong hydrochloric acid are 
 
 * Made by saturating ammonia with sulphurous acid, generated by heating 
 powdered charcoal and strong sulphuric acid in a flask. The mixture is 
 made of the consistency of cream, and the gas passed first through a wash- 
 b ottle containing water. 
 
ESTIMATION OF PHOSPHORIC ACID. 335 
 
 added to decompose any excess of ammonium sulphite 
 and the sulphurous acid driven off by passing carbonic acid 
 through the boiling solution. When the last trace of 
 sulphurous acid is driven off, sulphuretted hydrogen is 
 passed through the boiling solution to precipitate any 
 arsenic, the sulphide of arsenic filtered off, the beaker 
 placed in cold water, and when thoroughly cooled, dilute 
 ammonia added until a slight permanent green precipitate 
 of ferrous hydrate remained after stirring. Acetic acid 
 is added until the precipitate dissolved (a few drops 
 should be sufficient) the solution, diluted with hot water 
 to about 800 or 900 c.c. And if the precipitate is white, 
 a dilute solution of ferric chloride or bromine water is 
 added until it becomes red. If it is necessary to add 
 much ferric chloride a few drops of ammonic acetate are 
 also added to insure the decomposition of all the former 
 salt. The solution is then heated to boiling, boiled a 
 few minutes, filtered rapidly, and washed once with hot 
 water. The filtrate should pass through the filter perfectly 
 clear, and if the precipitate is red, any subsequent cloudi- 
 ness of the filtrate is of no consequence. If, however, 
 the filtrate passes the filter cloudy, it should be returned 
 to the main portion, a few drops more of ferric chloride 
 added, and the solution again boiled and filtered. The 
 precipitate of ferric phosphate, hydrate, and basic acetate 
 is dissolved in hot dilute hydrochloric acid (1 1) on 
 the filter, and the large beaker cleared of any adhering 
 precipitate with hydrochloric acid, received in a small 
 beaker, and evaporated nearly to dryness. Five to ten 
 grms. of citric acid are dissolved in the least possible 
 quantity of hot water, and filtered into a small beaker into 
 which are also filtered about 5 c.c. of magnesium mix- 
 ture,* and the whole is added to the solution of the 
 ferric phosphate. This solution is then neutralised by 
 ammonia and cooled. When perfectly cold, from one- 
 third to one-half its bulk of strong ammonia is added, 
 
 * Made by dissolving equal weights of sulphurous acid and ammonium 
 chloride in the least possible quantity of water, filtering, adding bulk of 
 strong ammonia, stirring occasionally, and allowing it to stand for some days 
 before using. 
 
336 COMPLETE ASSAY OF IRON ORES. 
 
 and the solution is stirred until the precipitate of ammo- 
 nium-magnesium phosphate begins to form. After standing 
 for some time the solution is stirred again, and the 
 stirring is repeated at intervals for an hour. It is allowed 
 to settle overnight, filtered on the asbestos felt in Dr. 
 Gooch's pierced crucible, washed with dilute ammonia 
 (1 3), dried on the pump, moistened with a drop or two 
 of ammonium nitrate in ammonia, dried and ignited until 
 the glow passes over the precipitate, cooled in a desiccator, 
 and weighed as magnesium phosphate. This precipitate 
 is then dissolved in dilute hot hydrochloric acid, the felt 
 is washed on the pump, and the crucible is heated to 
 redness and re-weighed. This weight, unless the precipitate 
 of magnesium phosphate contained silica, agrees perfectly 
 with the original weight. In the latter case, however, the 
 last weight of the crucible is subtracted from the weight 
 of the crucible with the precipitate to obtain the weight of 
 magnesium phosphate. The following table is used for cal- 
 culating the percentage of phosphoric acid or phosphorus, 
 instead of the factors 0-6396 or 0-2793. 
 
 When the amount of magnesium-ammonium phosphate 
 is large it is dissolved, boiled, and re -precipitated, as a 
 small amount of magnesia is liable to be carried down 
 mechanically with it. It is found necessary to avoid 
 heating the magnesium phosphate after the glow has 
 passed, as this salt is liable to attack the asbestos slightly 
 at high temperatures, causing a small loss in weight of 
 the felt on the subsequent solution of the precipitate in 
 dilute hydrochloric acid. For ores containing titanic acid 
 it is necessary to modify this process in several particu- 
 lars. It is found that under certain circumstances phos- 
 phoric acid combines with titanic acid, forming a salt 
 (possibly a phospho-titanate) very insoluble in hydrochloric 
 acid, so that in almost every instance phosphoric acid is 
 found in the insoluble silicious residue when this contains 
 titanic acid. Also, upon neutralising the main solution 
 after adding ammonium sulphide in ores containing titanic 
 acid, a fine white precipitate resembling barium sulphate 
 is formed, which usually remains after the subsequent 
 
ESTIMATION OF PHOSPHORIC ACID. 
 
 337 
 
 TABLE FOR PHOSPHORUS AND PHOSPHORIC ACID. 
 
 Mg 
 
 p 
 
 P 2 5 
 
 Mg 
 
 P 
 
 PA 
 
 Mg 
 
 p 
 
 P 3 5 
 
 1 
 
 0-003 
 
 0-006 
 
 35 
 
 0-098 
 
 0-224 
 
 68 
 
 0-190 
 
 0-434 
 
 2 
 
 0-005 
 
 0-013 
 
 36 
 
 0-100 
 
 0-230 
 
 69 
 
 0-193 
 
 0-441 
 
 3 
 
 0-008 
 
 0-019 
 
 37 
 
 0-103 
 
 0-237 
 
 70 
 
 0-195 
 
 0-448 
 
 4 
 
 0-011 
 
 0-026 
 
 38 
 
 0-106 
 
 0-243 
 
 71 
 
 0-198 
 
 0-454 
 
 5 
 
 0-014 
 
 0-032 
 
 39 
 
 0-109 
 
 0-249 
 
 72 
 
 0-201 
 
 0-460 
 
 6 
 
 0-017 
 
 0-038 
 
 40 
 
 0-112 
 
 0-256 
 
 73 
 
 0-204 
 
 0-467 
 
 7 
 
 0-019 
 
 0-045 
 
 41 
 
 0-114 
 
 0-262 
 
 74 
 
 0-207 
 
 0-473 
 
 8 
 
 0-022 
 
 0-051 
 
 42 
 
 0-117 
 
 0-269 
 
 75 
 
 0-209 
 
 0-479 
 
 9 
 
 0-025 
 
 0-057 
 
 43 
 
 0-120 
 
 0-275 
 
 76 
 
 0-212 
 
 0-486 
 
 10 
 
 0-028 
 
 0-064 
 
 44 
 
 0-123 
 
 0-281 
 
 77 
 
 0-215 
 
 0-492 
 
 11 
 
 0-031 
 
 0-070 
 
 45 
 
 0-126 
 
 0-287 
 
 78 
 
 0-218 
 
 0-499 
 
 12 
 
 0-033 
 
 0-077 
 
 46 
 
 0-128 
 
 0-294 
 
 79 
 
 0-221 
 
 0-505 
 
 13 
 
 0-036 
 
 0-083 
 
 47 
 
 0-131 
 
 0-300 
 
 80 
 
 0-223 
 
 0-512 
 
 14 
 
 0-039 
 
 0-089 
 
 48 
 
 0-134 
 
 0-307 
 
 81 
 
 0-226 
 
 0-518 
 
 15 0-042 
 
 0-096 
 
 49 
 
 0-137 
 
 0-313 
 
 82 
 
 0-229 
 
 0-524 
 
 16 
 
 0-045 
 
 0-102 
 
 50 
 
 0-139 
 
 0-319 
 
 83 
 
 0-232 
 
 0-531 
 
 17 
 
 0-047 
 
 0-108 
 
 51 
 
 0-142 
 
 0-326 
 
 84 
 
 0-235 
 
 0-537 
 
 18 
 
 0-050 
 
 0-115 
 
 52 
 
 0-145 
 
 0-332 
 
 85 
 
 0-237 
 
 0-544 
 
 19 
 
 0-053 
 
 0-121 
 
 53 
 
 0-148 
 
 0-339 
 
 86 
 
 0-240 
 
 0-550 
 
 20 
 
 0-056 
 
 0-128 
 
 54 
 
 0-151 
 
 0-345 
 
 87 
 
 0-243 
 
 0-556 
 
 21 
 
 0-059 
 
 0-134 
 
 55 
 
 0-154 
 
 0-352 
 
 88 
 
 0-246 
 
 0-563 
 
 22 
 
 0-061 
 
 0-141 
 
 56 
 
 0-156 
 
 0-358 
 
 89 
 
 0-248 
 
 0-569 
 
 23 
 
 0-064 
 
 0-147 
 
 57 
 
 0-159 
 
 0-364 
 
 90 
 
 0-251 
 
 0-576 
 
 24 
 
 0-067 
 
 0-153 
 
 58 
 
 0-162 
 
 0-371 
 
 91 
 
 0-254 
 
 0-582 
 
 25 
 
 0-070 
 
 0-159 
 
 59 
 
 0-165 
 
 0-377 
 
 92 
 
 0-257 
 
 0-588 
 
 26 
 
 0-073 
 
 0-166 
 
 60 
 
 0-167 
 
 0-384 
 
 93 
 
 0-259 
 
 0-595 
 
 27 
 
 0-075 
 
 0-173 
 
 61 
 
 0-170 
 
 0-390 
 
 94 
 
 0-262 
 
 0-601 
 
 28 
 
 0-078 
 
 0-179 
 
 62 
 
 0-173 
 
 0-396 
 
 95 
 
 0-265 
 
 0-607 
 
 29 
 
 0-081 
 
 0-185 
 
 63 
 
 0-176 
 
 0-403 
 
 96 
 
 0-268 
 
 0-614 
 
 30 
 
 0-084 
 
 0-192 
 
 64 
 
 0-179 
 
 0-409 
 
 97 
 
 0-271 
 
 0-620 
 
 31 
 
 0-086 
 
 0-198 
 
 65 
 
 0-181 
 
 0-416 
 
 98 
 
 0-274 
 
 0-627 
 
 32 
 
 0-089 
 
 0-204 
 
 66 
 
 0-184 
 
 0-422 
 
 99 
 
 0-276 
 
 0-633 
 
 33 
 
 0-092 
 
 0-211 
 
 67 
 
 0-187 
 
 0-428 
 
 100 
 
 0-279 
 
 0-639 
 
 34 
 
 0-095 
 
 0-217 
 
 
 
 
 
 
 
 addition of hydrochloric acid, and consists of titanic 
 acid with some phosphoric acid. If, after dissolving 
 the precipitate of ferric phosphate, &c., thrown down by 
 ammonic acetate, the solution is allowed to run to dry- 
 ness, there is found after re-solution in hydrochloric 
 acid a granular white or yellowish residue, absolutely 
 insoluble in hydrochloric acid, which consists essentially 
 of titanic and phosphoric acid. This reaction affords a 
 very delicate test for titanic acid, for if the residue obtained 
 in this latter case be collected on a small filter, dried, 
 ignited, and fused with sodium carbonate, and the fusion 
 treated with hot water, a residue of sodium titanate 
 will remain, which is insoluble in water. This residue 
 
 z 
 
338 COMPLETE ASSAY OF IRON ORES. 
 
 collected on a small filter, dissolved in a little hot dilute 
 hydrochloric acid, and treated with zinc in a test-tube, 
 gives the very characteristic violet colour due to titanic 
 oxide. When titanic oxide is found it is necessary to 
 fuse the insoluble silicious residue, the residue left in the 
 filter from the solution of the precipitate by ammonic 
 acetate, and the residue left from the re-solution of this 
 latter, after it has run to dryness, with sodium carbonate. 
 The fused mass is treated with hot water and filtered ; the 
 filtrate containing, besides silica and alumina and sodium 
 carbonate, all the phosphoric acid, while the titanic acid 
 remains on the filter as sodic titanate, insoluble in water. 
 Potassium salts should not be used, as potassic titanate is 
 decomposed by water. The filtrate is acidulated with 
 hydrochloric acid, .evaporated to dryness, dissolved in 
 water with a little hydrochloric acid, filtered, a few drops 
 of solution of ferric chloride added, and the ferric phos- 
 phate precipitated by ammonic acetate. This precipitate 
 is filtered, washed, dissolved in hydrochloric acid, and 
 the phosphoric acid estimated by magnesium mixture 
 as usual. 
 
 Sulphur and Iron. 
 
 One grm. of ore is fused with 10 to 12 grms. of 
 sodium carbonate and a little nitre, the fused mass is 
 run well up on the sides of the crucible, and the cru- 
 cible is chilled. The mass is then detached from the 
 crucible and transferred to a tall beaker, or the crucible 
 with its contents is placed in the beaker and treated 
 with boiling water. When the fused mass is entirely 
 disintegrated (the crucible, if placed in the beaker, having 
 been washed off and removed) the ferric oxide is allowed 
 to settle. If the solution is coloured by alkaline man- 
 ganate a few drops of alcohol are added, and the solution 
 is allowed to stand until the colour disappears. If the 
 solution after the disappearance of the colour due to the 
 manganate is yellow, it is an indication of the presence 
 of chromium in the ore. It is then decanted through a 
 filter and washed twice with hot water by decantation. 
 
SULPHUR AND IRON. 339 
 
 The filter is washed once or twice with hot water, 
 and the filtrate is acidulated with hydrochloric acid, 
 evaporated to dryness, re-dissolved in hot water with 
 a few drops of hydrochloric acid, filtered from the 
 silica, and the sulphuric acid precipitated by barium 
 chloride in the boiling filtrate, After standing over- 
 night in a warm place the barium sulphate is fil- 
 tered on the felt, washed thoroughly with hot water 
 (and ammonium acetate if it is large), and weighed 
 as barium sulphate, which multiplied by 0-1373 gives 
 the sulphur. 
 
 The crucible in which the fusion was made is treated 
 with hydrochloric acid, and when the adhering ferric oxide 
 is dissolved a little hot water is added, and the whole 
 is poured on the filter to dissolve any ferric oxide which 
 might have been in suspension in the solution of the fused 
 mass. This is allowed to run into the beaker containing 
 the ferric oxide, the crucible is rinsed, and the filter 
 is washed with hot water. The whole is evaporated 
 to dryness to render silicic acid insoluble, re-dissolved 
 in hydrochloric acid, and transferred to a small flask 
 of about 50 c.c. capacity. A small funnel is put in 
 the neck of the flask, and 3 grms. of granulated zinc in 
 small lumps are added carefully to the solution. By 
 heating the solution and adding a few drops of hydro- 
 chloric acid from time to time the ferric chloride is soon 
 reduced to ferrous, which is shown by the solution be- 
 coming colourless. After washing down the neck of the 
 flask and the funnel with a fine jet of water, if the addition 
 of a few drops of hydrochloric acid fails to impart any 
 colour to the solution, the reduction is considered com- 
 plete. Fifteen c.c. of dilute sulphuric acid (1 1) are 
 added little by little to the solution, and when the zinc 
 has all dissolved the neck of the flask is filled nearly 
 to the top with hot water. The flask is then placed 
 in cold water, and when the solution is thoroughly 
 cooled it is washed out into an oblong white dish of 
 about 1,500 c.c. capacity and diluted to about one 
 litre. A solution of potassium permanganate is run 
 
 z 2 
 
340 
 
 COMPLETE ASSAY OF IRON ORES. 
 
 FIG. 91. 
 
 in from a burette, the representation of which is given in 
 fig. 91. 
 
 This form of burette, the invention of Mr. Thomas H. 
 
 Garrett, is the most satis- 
 factory one I have ever used. 
 The solution of permanga- 
 nate is carefully standard- 
 ised by means of ferric chlo- 
 ride of known strength, made 
 by dissolving wrought iron 
 of known composition in ni- 
 tric acid, driving off the nitric 
 acid by repeated evapora- 
 tions with hydrochloric acid, 
 diluting, filtering into a clean 
 glass-stoppered bottle, and 
 estimating the ferric oxide 
 gravimetrically in a weighed 
 amount of the solution. The 
 strength of the ferric chlo- 
 ride solution being thus ac- 
 curately known, a portion 
 is weighed out into one of 
 the little flasks previously 
 mentioned, reduced with zinc, 
 sulphuric acid added, and 
 the strength of the perman- 
 ganate solution necessary to 
 colour 3 grms. of zinc treated 
 with hydrochloric acid and 
 sulphuric acid, and diluted to 
 
 the same volume as the solution of the ore, is subtracted 
 before calculating the value of the permanganate solution , 
 and also from the amount of permanganate solution re- 
 quired for the ore before calculating the percentage of 
 iron. By using sulphuric acid in addition to the hydro- 
 chloric acid as described above it is found that the end 
 reaction with potassium permanganate is as sharp as if no 
 hydrochloric acid is used, and the results obtained are 
 
SILICA, FERRIC OXIDE, ALUMINA, ETC. 341 
 
 in all cases as accurate as could be desired. For many 
 ores it is possible to dissolve the finely ground sample 
 directly in the flask in hydrochloric acid without previous 
 fusion, but for those containing ferrous silicates or pyrites 
 it is generally necessary to resort to fusion for the accu- 
 rate estimation of the total iron. 
 
 Silica, Ferric Oxide, Alumina, Manganese, Lime, and 
 Magnesia. 
 
 One gramme is dissolved in hydrochloric acid, evapo- 
 rated to dryness, re-dissolved in dilute hydrochloric acid, 
 and evaporated a second time to render the silica in- 
 soluble. It is then re-dissolved in hydrochloric acid and 
 water (1 3), filtered on a small ashless filter, washed, 
 dried, ignited, and weighed as insoluble silicious matter. 
 This is fused with five times its weight of dry sodium 
 carbonate, treated with hot water, and washed into a 
 platinum dish, the crucible treated with acid and carefully 
 rinsed into the dish, the whole acidulated with hydro- 
 chloric acid, evaporated to dryness, treated with water and 
 a little hydrochloric acid, and evaporated to dryness. 
 
 The mass is then treated with hydrochloric acid and 
 water (] 5), heated to boiling, and filtered on a small 
 ashless filter. The filtrate is dried, ignited, and weighed. 
 The silica is treated in the crucible with hydrofluoric 
 acid * and sulphuric acid, evaporated to dryness, ignited, 
 and weighed. The difference between this and the weight 
 of the precipitate is silica. Any residue obtained by 
 the treatment with hydrofluoric acid and sulphuric acid 
 is examined. It might consist of barium sulphate or 
 titanic acid if either of these exists in the ore, or of a 
 little alumina or ferric oxide or of a small amount of 
 sodium sulphate if the silica is not carefully washed. 
 The filtrate from the silica is treated in a platinum dish 
 with ammonia, boiled until it smells but faintly of 
 
 * The hydrofluoric acid is re-distilled with the addition of a little sul- 
 phuric acid and potassium permanganate from a platinum still, and collected in 
 a platinum bottle, as the crude hydrofluoric acid always contains ferric oxide, 
 besides various sulphur compounds. 
 
342 COMPLETE ASSAY OF IRON ORES. 
 
 ammonia, filtered, washed, dried, ignited, and weighed as 
 alumina with or without a tinge of ferric oxide. To 
 the filtrate, ammonium oxalate is added, the solution is 
 boiled and allowed to stand overnight, filtered, washed, 
 and ignited at a high temperature, and weighed as lime. 
 
 This filtrate is evaporated down, sodium and am- 
 monium phosphate are added, and the solution is well 
 stirred to precipitate the magnesium-ammonium phos- 
 phate. After standing overnight it is filtered, washed, 
 ignited, and weighed as magnesium phosphate. Whence 
 111 : 40 = weight of magnesium phosphate to weight of 
 magnesia. The sum of the weights of the silica, alumina, 
 lime, and magnesia should about equal that of the in- 
 soluble silicious matter. When there is a deficiency, 
 alkalies are looked for in another portion. If there is 
 an excess, and the precipitated alumina is red, iron in 
 insoluble matter is estimated. 
 
 The filtrate from the insoluble silicious matter is nearly 
 neutralised with sodium carbonate solution, 2 grammes 
 of sodium acetate are added, and the whole after being 
 diluted to about 700 c.c. with hot water is boiled, and 
 the precipitate ferric oxide, &c., is filtered on a washed 
 filter. The precipitate is washed two or three times on 
 the filter, and then transferred back to the beaker with 
 a platinum spatula, the filter is washed with hydro- 
 chloric acid and finally with water, the whole being re- 
 ceived in the beaker with the mass of the precipitate. 
 Sufficient acid is added to dissolve the precipitate, and 
 the operation is repeated, the filtrates being added to- 
 gether. The solution of the precipitated ferric oxide, &c.,, 
 is evaporated to dryness to render insoluble any silica 
 from the reagents ; re-dissolved in dilute acid, and filtered 
 into a large platinum dish ; the solution is boiled, am- 
 monia is added, and the precipitate is collected on an 
 ashless filter, washed thoroughly on the pump with hot 
 water, dried, ignited (finally over the blast), and weighed 
 as iron and aluminium phosphate. The filtrates from the 
 acetate precipitations are evaporated down to about 300 
 c.c., 2 or 3 grammes of sodic acetate are added, the- 
 
SILICA, FEEKIC OXIDE, ALUMIXA, ETC. 343 
 
 solution is heated to boiling, and sulphuretted hydrogen 
 is passed through to precipitate any copper, nickel, 
 cobalt, and zinc. 
 
 The precipitate by sulphuretted hydrogen is filtered 
 off', and after all smell of sulphuretted hydrogen has 
 been boiled off, bromine water is added to the solution. 
 When the precipitated manganic oxide has collected at 
 a gentle heat, and while the solution is still coloured by 
 bromine, it is boiled until colourless, filtered, washed 
 several times, and the manganic oxide is dissolved on 
 the filter in hydrochloric acid with the addition of solu- 
 tion of sulphurous acid, which causes its very rapid solu- 
 tion. This solution is evaporated to dryness, re-dissolved 
 in hydrochloric acid and water ; a slight excess of am- 
 monia is added, nearly all smell of ammonia is boiled 
 off, the solution is filtered, any slight precipitate of 
 ferric oxide, &c., is re-dissolved and re-precipitated and 
 filtered as before. The filtrates are added together, and 
 excess of sodium- ammonia phosphate is added with 
 enough hydrochloric acid to render the solution decidedly 
 acid, and after boiling for some time an excess of am- 
 monia is added to precipitate the manganese-ammonium 
 phosphate. This is boiled until the precipitate becomes 
 crystalline and the solution smells but slightly of am- 
 monia, when it is filtered, washed with cold water, ignited, 
 and weighed as manganese phosphate, which multiplied 
 by 0-5 gives oxide of manganese. 
 
 To the filtrate from the precipitate of manganic oxide 
 by bromine is added the ammoniacal filtrate from the 
 final precipitation of the iron, alumina, and phosphoric 
 acid ; the whole is evaporated down to about 400 c.c., 
 and the lime and magnesia are precipitated as described 
 in the analysis of the insoluble silicious matter. There 
 are several sources of error which it is found very 
 necessary to guard against in the above analysis namely, 
 the contamination of the distilled water by silica when it 
 is boiled in glass, the strong action of ammoniacal solu- 
 tions on the beakers, and the presence of silica, alumina, 
 ferric oxide, lime, and magnesia in many reagents, and 
 
344 COMPLETE ASSAY OF IRON OKES. 
 
 especially in sodium carbonate. Distilled water which has 
 been heated overnight in a Bohemian flask on the sand- 
 bath is found to contain 52 milligrammes of solid residue 
 to the litre, 26 milligrammes of which is silica ; and 
 ferric oxide precipitated from a solution to which water 
 boiled in a flask from two to six hours has been added, 
 contains after ignition as much as 8 per cent, of silica. 
 To avoid this tin-lined copper flasks are used for heat- 
 ing distilled water, and to avoid the error due to the 
 action of ammonia salts and ammoniacal solutions on the 
 beakers, all the precipitations and evaporations are made 
 in platinum. We were fortunate enough to obtain some 
 remarkably pure sodium carbonate, containing only 2 
 milligrammes of silica, 1^ milligramme of alumina, 2 
 milligrammes of lime, and 2 milligrammes of magnesia to 
 the 100 grammes from Messrs. Powers and Weightman, 
 Philadelphia. 
 
 Nickel, Cobalt, and Zinc. 
 
 Three grammes of ore are treated as if for the esti- 
 mation of manganese, and the sulphides of copper, 
 nickel, cobalt, and zinc are precipitated from the boiling 
 solution of the acetates by sulphuretted hydrogen. This 
 precipitate is collected on a small filter, washed with 
 sulphuretted hydrogen water containing a little acetic 
 acid, dried, ignited in a porcelain crucible, and transferred 
 to small beaker. It is then treated with hydrochloric 
 acid and a little nitric acid, evaporated to dryness, re- 
 dissolved in hydrochloric acid, diluted, boiled, and the 
 copper precipitated by sulphuretted hydrogen. The 
 copper sulphide is filtered off and washed with hot 
 water, the filtrate containing the cobalt, nickel, and zinc 
 is evaporated to dryness, re-dissolved in a few drops 
 of water containing not more than two or three drops of 
 hydrochloric acid, and an excess of potassium nitrate 
 acidulated with acetic acid added. After standing for a 
 day or two the potassic-cobaltic nitrite is filtered off, 
 washed once or twice with a strong solution of potassic 
 acetate, and then, after removing the filtrate, with alcohol 
 
NICKEL, COBALT, AND ZINC. 345 
 
 to displace the alkaline salts. The precipitate is then 
 ignited very carefully in a porcelain crucible, treated with 
 sulphuric acid to decompose all the nitrate, made alkaline 
 with ammonia, filtered, and the cobalt is precipitated by 
 the battery ; or the ignited precipitate is dissolved in a few 
 drops of hydrochloric acid, transferred to a small beaker, 
 dilated, and any alumina and iron present is precipi- 
 tated by boiling for several hours with an excess of sodic 
 acetate. The precipitate of alumina and iron is filtered 
 off, the filtrate is made alkaline with ammonia, and the 
 cobalt is precipitated as sulphide by ammonium sul- 
 phide. The precipitate is allowed to settle, filtered off 
 on a small ashless filter, washed, dried, ignited, and weighed 
 as cobaltous sulphide, or treated with sulphuric acid and 
 zinc, and weighed as cobaltic sulphide, which multiplied 
 by 0*4872 gives cobalt oxide. 
 
 The filtrate from the potassium- cobaltic nitrite is 
 acidulated strongly with hydrochloric acid and heated to 
 decompose all the nitrite, and the nickel oxide is preci- 
 pitated by an excess of soda or potash, filtered, and the 
 filtrate tested for zinc oxide with ammonium sulphide. 
 If any zinc sulphide is found, the nickel oxide is re- 
 dissolved and precipitated as before, filtered, and the 
 filtrate and washings are added to the first filtrate. The 
 zinc sulphide is allowed to settle, filtered, washed with 
 water containing sulphide of ammonium, re-dissolved in 
 hydrochloric acid, and evaporated to dryness. This 
 is treated with dilute hydrochloric acid, filtered, and 
 the zinc is precipitated by solution of sodium carbonate, 
 filtered, washed, dried, ignited, and weighed as zinc oxide. 
 
 The precipitate of nickel oxide is dissolved on the 
 filter in hydrochloric acid, the filtrate run to dryness with 
 sulphuric acid, diluted, excess of ammonia is added, 
 filtered, and the nickel precipitated by the battery ; or 
 the hydrochloric acid solution is evaporated to dryness, 
 dissolved in a drop or two of hydrochloric acid, diluted, 
 and boiled with an excess of sodium acetate. Any pre- 
 cipitate of alumina, &c., is filtered off, excess of sulphide 
 of ammonium is added, then an excess of acetic acid 
 
346 
 
 COMPLETE ASSAY OF IRON ORES. 
 
 and sulphuretted hydrogen is passed through the boil- 
 ing solution. The precipitated nickelic sulphide and 
 sulphur are collected on a small ashless filter, dried, 
 ignited, heated with a little ammonic carbonate, and 
 weighed as nickelous sulphide. 
 
 Estimation of Ferrous Oxide. 
 
 One gramme of finely ground ore is weighed into a 
 flask A (fig. 92) of about 100 c.c. capacity, fitted with a 
 
 FIG. 92. 
 
 rubber stopper, through which pass two glass tubes as 
 shown in the cut. Dry carbonic acid is passed in through 
 the tube B to expel the air through the tubes c and D, 
 the latter dipping beneath the surface of the water in 
 the beaker E, and when this is accomplished the stop- 
 per is removed, 15 c.c. of strong hydrochloric acid is 
 quickly poured in, the stopper is replaced, and the ore 
 is dissolved with the aid of heat, the current of carbonic 
 acid being continued uninterruptedly. When the solution 
 
ESTIMATION OF FEEEOUS OXIDE. 347 
 
 of the ore is accomplished the source of heat is re- 
 moved, and, the current of carbonic acid being temporarily 
 stopped, the water in the beaker B is allowed to run 
 back into the flask A until the latter is nearly filled, 
 when the current of carbonic acid is turned on again, 
 and allowed to continue until the solution in the flask is 
 thoroughly cooled. This is accomplished by removing 
 the tripod c, placing a dish nearly filled with cold water 
 under the flask, and replacing the tripod. The solution 
 is then washed out into the dish used for titrating, into 
 which 3 grms. of zinc dissolved in 15 c.c. of the dilute 
 sulphuric acid (1 1) are previously poured, and the whole 
 is diluted to 1 litre. The amount of ferrous oxide is 
 then estimated by the potassic permanganate solution. 
 The amount of solution required to colour about 1 grm. 
 of ferric chloride diluted to the same bulk, and containing 
 the same amount of sulphuric acid, is subtracted from 
 the amount required for the titration before calculating 
 the amount of ferrous oxide. 
 
 Ferrous Oxide in Insoluble Silicious Matter. 
 
 When the insoluble silicious matter contains iron in 
 the ferrous condition for instance, in the form of epidote 
 1 grm. is treated with hydrochloric acid diluted, and 
 the insoluble matter is collected on the felt in a pierced 
 crucible, carefully dried, transferred with the felt to an 
 ordinary platinum crucible, and treated in the apparatus 
 of fig. 93, with hydrofluoric acid and hydrochloric acid in a 
 current of carbonic acid. When entirely decomposed it 
 is allowed to cool, the current of carbonic acid being 
 kept up, and then transferred quickly to a dish containing 
 about 1 litre of water and 15 c.c. of the dilute sulphuric 
 acid and zinc ; the crucible is washed out, and the amount 
 of ferrous oxide is estimated by the standard solution 
 of potassium permanganate. 
 
348 
 
 COMPLETE ASSAY OF IRON ORES. 
 
 Sulphuric Acid. 
 
 Sulphuric acid may exist in the ore in the form of 
 barium sulphate, calcium sulphate, or as sulphates of iron, 
 
 FIG. 93. 
 
 &c., formed in the roasting or weathering of an ore con- 
 taining pyrites. When it exists as barium sulphate it is 
 found in the analysis of the insoluble silicious matter in 
 the residue remaining after the treatment of the silica with 
 hydrofluoric acid. In this case the residue is fused with 
 sodium carbonate, treated with hot water, and filtered. 
 The filtrate is acidulated with hydrochloric acid, and 
 the sulphuric acid is precipitated by barium chloride 
 and estimated as barium sulphate, which multiplied by 
 0*3433 gives sulphuric anhydride. 
 
 The barium carbonate on the filter is dissolved in 
 dilute hydrochloric acid, sulphuric acid is added, and 
 the barium sulphate is finally weighed, which multiplied 
 
SULPHUEIC ACID. 34$ 
 
 by 0-6567 gives baryta. In the other cases 10 grammes 
 of the ore finely ground are treated with water contain- 
 ing a little hydrochloric acid, filtered, the sulphuric acid 
 is precipitated in the filtrate as barium sulphate, and the 
 weight is determined with the usual precautions. 
 
 Alumina. 
 
 The total amount of the soluble ferric oxide, alumina, 
 and phosphoric acid is added to the ferric oxide and 
 alumina found in the insoluble silicious matter, and from 
 this is subtracted the phosphoric acid which gives the 
 total iron and alumina. From this is subtracted the 
 ferric oxide found by titration ; the difference is alumina. 
 The accuracy of this result depends of course upon that 
 of the volumetric estimation of the ferric oxide. The 
 comparison of many analyses shows that the error due 
 to this need never exceed a few hundredths of 1 per cent., 
 and no direct method is found to yield results equally 
 accurate or concordant. 
 
 Calculation of the Analysis. 
 
 The sulphur in the sulphuric acid found as such is 
 subtracted from the total sulphur ; so much as is neces- 
 sary to form the sulphides of copper, nickel, cobalt, zinc, 
 and antimony with the amounts of these elements in the 
 ore, and supposed to exist in this condition, is sub- 
 tracted from this, and the remaining sulphur is calcu- 
 lated to iron disulphide. The amount of iron required 
 for this is calculated to ferric oxide, and subtracted from 
 the ferric oxide found by titration. From this is taken 
 the amount of ferric oxide calculated from the ferrous 
 oxide found in the estimation of ferrous oxide, and the 
 result is the amount of ferric oxide existing in this state 
 in the ore. 
 
 Carbonic Acid. 
 
 Three grammes of ore are weighed into the flask A 
 of the apparatus (fig. 94), and the connections are made 
 as there shown ; 10 c.c. of strong sulphuric acid are 
 
350 
 
 COMPLETE ASSAY OF IRON ORES. 
 
 added to 65 c.c. of water and poured into the bulb-tube 
 B, the stopcock c having first been closed. After the 
 
 potash bulb and the drying-tube are weighed they are 
 attached to the apparatus, the tube s is filled with fused 
 calcium chloride, being added to prevent the drying-tube 
 R from, absorbing moisture from the atmosphere. The 
 
WATEE AND CARBON. 351 
 
 TJ-tube o is empty, p is filled with pumice saturated 
 with anhydrous copper sulphate, and Q with dried calcium 
 chloride. When the connections are all proved to be 
 tight, N is fitted into the neck of the bulb-tube with a 
 piece of rubber tubing, and the acid in the bulb is 
 allowed to run into the flask very slowly, and when it is 
 all in, a slow current of air is forced through the appa- 
 ratus by means of the bottles L L. The air is freed from 
 carbonic acid by potassic hydrate in the tube M. As soon 
 as the current of air is started, the flask A is heated 
 gradually, and finally the solution is boiled until the 
 bend of the tube o is filled with condensed water. It 
 is then allowed to cool while the current of air is con- 
 tinued. The potash bulb and the drying-tube are de- 
 tached and weighed with the usual precautions ; the 
 increase of weight is of course the carbonic acid due to 
 the carbonates contained in the ore. 
 
 Water and Carbon in Carbonaceous Matter. 
 
 Many ores besides black-bands contain carbon ; for in- 
 stance, several ores from New Jersey contain graphite, 
 and nearly all limonites and magnetites contain carbon in 
 organic matter, probably from the organic acids which 
 originally held the iron in solution. As pyrites is also 
 of common occurrence in such ores it is necessary to 
 devise some method by which the water of composition 
 and the carbon could be estimated in the presence of 
 pyrites. In attempting to estimate the water by heating 
 the ore in a current of air, some sulphuric acid is always 
 formed and driven over into the drying-tube, and some 
 organic matter is certain to be dissolved if the ore is 
 treated with acid for the estimation of the carbon. It 
 is found by careful experiments that when a carbonate 
 of any kind is fused with potassium anhydrous chromate in 
 excess the carbonic acid is all expelled together with 
 any water present, while pyrites treated in the same way 
 is oxidised with formation of potassium sulphate, which 
 is not decomposed even at a bright red heat. 
 
352 
 
 COMPLETE ASSAY OF IRON OEES. 
 
 A very practical and easy way to apply this in the 
 treatment of ores is afforded by the use of Dr. Gooch's 
 tubulated crucible* (fig. 95), and the process is as 
 follows : 
 
 For ores containing much water or carbonic acid I 
 
 FIG. 95. 
 
 HES 
 
 gramme, and for others 3 grammes are weighed into the 
 crucible A of fig. 96, and carefully mixed, by means of a 
 rod or wire, with 7 grammes of potassium bichromate, 
 which has been previously fused and powdered, the cap 
 B is adjusted, and the whole is placed in the air-bath 
 and heated to 100 for a short time. The crucible is 
 then placed on the platinum triangle c, and connected by 
 means of a cork with the weighed tube D, containing 
 dried calcium chloride, and the weighed potash bulb E 
 and drying-tube F are attached, the latter guarded by 
 the calcium chloride tube G, as seen in the sketch. 
 
 The cap for the crucible consists of a conical cover H 
 drawn out vertically into a tube i, into which is burned 
 a horizontal tube J, of the same diameter. Through the 
 top of the tube I passes the tube K to a distance of half an 
 inch below the bottom of the cap, the end being slightly 
 bent ; K is burned into I at its point of entry a, sealing i 
 
 * ' Chemical News,' vol. xiii. p. 326. 
 
WATER AND CARBON. 
 
 353 
 
 at this point. A straight glass tube M, drawn tapering, 
 is fused to the platinum tube j at b and another piece, N, 
 
 is bent at a right angle to the platinum K at c. A piece 
 of rubber tubing, closed with a piece of glass rod at one 
 end, is drawn over the end of the tube N, the space 
 around the cap in the flange d is filled with small 
 
 A A 
 
354 COMPLETE ASSAY OF IRON ORES. 
 
 pieces of neutral anhydrous -sodic tungstate, which are 
 fused by means of a blowpipe flame, making an air-tight 
 joint. The mixture in the crucible is kept cool during 
 this operation by dipping the end in a beaker of cold 
 water. The expansion of the air in the apparatus during 
 the heating causes it to bubble through the potash bulb, 
 and the reflux of the solution in the bulb as the apparatus 
 cools is a good index of the tightness of the joints. 
 When the joints are shown to be all tight the cap is 
 removed from N, and the bottle L is connected with N, 
 as shown in the sketch. A slow current of air, freed from 
 carbonic acid and water by passing through the tube Q, 
 containing potash and calcium chloride, is then started 
 through the apparatus, and the crucible is heated very 
 gradually and cautiously by the burner o. As the steam 
 is gradually driven into the drying-tube it is not allowed 
 to condense at y, but is driven forward into the calcium 
 chloride by the heat of a small alcohol lamp, applied 
 to the drying-tube at this point. When the greater part 
 of the water has been thus driven over, the crucible 
 is heated by a horizontal flame from the blast-lamp p, 
 above the level of the mixture to prevent the latter from 
 frothing up and stopping the end of the tube K. The 
 bottom of the crucible is gradually heated to a dull red, 
 and allowed to remain at this temperature for about fifteen 
 minutes, when the lamps are turned out and the whole 
 is allowed to cool in the current of dried air from the 
 bottle L. The tubes are then re-weighed ; the increase 
 in the weight of D is the weight of the water of com- 
 position in the ore, and that of the potash bulb and drying- 
 tube the weight of the total carbonic acid. This latter, of 
 course, is the sum of the carbonic acid, due to the car- 
 bon of the carbonaceous matter or graphite, and of that 
 due to the carbonic acid in the carbonates. The amount 
 of the latter (previously estimated) is subtracted from 
 the total carbonic acid, and the difference multiplied by 
 0-2626 gives the carbon in carbonaceous matter or the 
 graphite. 
 
ALKALIES. 355 
 
 Alkalies. 
 
 The alkalies are generally confined to the insoluble 
 silicious matter in the ores, but are occasionally found 
 in the portion soluble in hydrochloric acid. In the former 
 case 5 or 10 grammes of the ore are digested in hydro- 
 chloric acid until only the silicates remained undissolved, 
 diluted, and filtered. The insoluble residue is washed, 
 dried, ignited, and treated in the crucible with hydrofluoric 
 and sulphuric acids until it is entirely decomposed ; it 
 is then evaporated down until all the hydrofluoric acid 
 and nearly all the sulphuric acid has volatilised. It is then 
 treated with hot water and a little hydrochloric acid if 
 necessary ; the clear solution is transferred to a small 
 platinum dish, the alumina and ferric oxide are pre- 
 cipitated by ammonia, and the solution is boiled until 
 nearly all ammonia is driven off. It is then filtered into 
 another platinum dish, evaporated to dryness, and heated 
 until all the ammonium salts are volatilised. The residue 
 is treated with a little water, a few drops of ammonium 
 oxalate, and excess of ammonia, and the solution is 
 boiled and filtered. The filtrate is evaporated to dry- 
 ness, and the residue is heated to dull redness. It is then 
 treated with water, filtered, and an excess of barium 
 hydrate * is added to precipitate the sulphuric acid and 
 magnesia. The solution is boiled and filtered from the 
 barium sulphate, and the filtrate is evaporated to dry- 
 ness after the addition of a little ammonium carbonate 
 and hydrate. The residue is treated with a little water, 
 filtered to get rid of the barium carbonate, and the filtrate 
 is evaporated to dryness and heated carefully to vola- 
 tilise any ammonium salts. The residue is treated with 
 water, filtered into a small weighed platinum dish ; the 
 filtrate is acidulated with hydrochloric acid, evaporated 
 to dryness, heated carefully to a very low red, and weighed 
 as quickly as possible as alkaline chlorides. The alkaline 
 chlorides are then dissolved in water, any residue is 
 
 * Made by treating ordinary barium carbonate with tolerably strong nitric 
 acid, and washing the barium nitrate with nitric acid. The barium nitrate is 
 dried and fused until all the nitrous acid is driven off. 
 
 A A 2 
 
356 COMPLETE ASSAY OF IRON ORES. 
 
 filtered off, weighed, and its weight subtracted from the 
 weight of the chloride, an excess of platinum chloride is 
 added, and the solution is evaporated on the water-bath 
 until the syrupy liquid solidifies on cooling. The residue 
 is treated with alcohol, 95 per cent, filtered on the felt, 
 washed with 95 per cent, alcohol, dried at 120 C., and 
 weighed as platinum-potassium chloride, which multiplied 
 by 0*1931 gives potassium oxide. The weight of the 
 platinum-potassium chloride, multiplied by 0-3056 to re- 
 duce to potassium chloride, is subtracted from the weight 
 of the alkaline chlorides, and the difference (sodium 
 chloride), multiplied by 0-5299, gives sodium oxide. 
 
 The filtrate from the platinum-potassium chloride is 
 evaporated to dryness in a platinum dish with the addition 
 of a little oxalic acid to decompose all the platinum 
 chloride ; the residue is treated with water, filtered, and 
 any magnesia is precipitated by microcosmic salt and 
 weighed as magnesium phosphate, which multiplied by 
 0- 4*2 8 gives the amount of magnesium chloride to be sub- 
 tracted from the weight of chlorides to get the amount of 
 alkaline chlorides. 
 
 For the estimation of alkalies in the portion of the ore 
 soluble in hydrochloric acid 1 gramme is treated with 
 hydrochloric acid, diluted, filtered into a platinum dish, 
 and heated to boiling ; a slight excess of ammonia is 
 added, and the whole is evaporated to dryness to render 
 the ferric oxide very granular and easy to wash. It is 
 then treated with water, a drop or two of ammonia is 
 added, and the ferric oxide, &c., filtered off and washed. 
 The filtrate, after the addition of a drop or two of sulphuric 
 acid, is evaporated to dryness, and the ammoniacal salts 
 are driven off by ignition. The residue is treated with 
 water, a little ammonium oxalate is added to precipitate 
 the lime, and the alkalies are estimated, as in the former 
 case, after the precipitation of the calcium oxalate. 
 
 Copper, Lead, Arsenic, and Antimony. 
 
 Ten grammes of the very finely ground ore are treated 
 with hydrochloric acid with the addition of potassium 
 
COPPER, LEAD, ARSENIC, AND ANTIMONY. 
 
 357 
 
 chlorate until only the silicious residue remains unacted 
 upon. It is then diluted, filtered, reduced with ammo- 
 nium sulphite, all the excess of sulphurous acid is boiled 
 off, and sulphuretted hydrogen is passed through to satu- 
 ration. The precipitate is allowed to settle in a warm 
 place, and when the smell of sulphuretted hydrogen has 
 nearly passed off, it is filtered on the pump, washed, 
 and treated on the filter with potassium sulphide to dis- 
 solve out the sulphides of arsenic and antimony. The 
 filter containing the sulphides of copper and lead is 
 dried and ignited in a porcelain crucible. The ignited 
 sulphides are transferred to a small beaker and treated 
 with hydrochloric acid and nitric acid, excess of sulphuric 
 acid added, and the whole is evaporated down until sul- 
 phuric acid begins to volatilise. Water is then added, 
 and, if any lead sulphate separates out, an equal bulk of 
 alcohol, and the whole is allowed to stand some hours, 
 when the lead sulphate is filtered off, dissolved in basic 
 ammonium tartrate or acetate, and the lead sulphide pre- 
 cipitated by sulphuretted hydrogen is filtered, washed, 
 and ignited. It is treated in the crucible (porcelain) 
 with nitric and sulphuric acids, and finally ignited and 
 weighed as lead sulphate, which multiplied by 0-6832 gives 
 lead. The solution containing the copper is evaporated 
 down to drive off the alcohol, washed 
 out into a platinum crucible, and the 
 copper is precipitated on the cru- 
 cible by the battery in the apparatus 
 shown in fig. 97, washed with water 
 and alcohol, and weighed as copper. 
 
 The potassium sulphide solution 
 containing the sulphides of arsenic 
 and antimony is acidulated with 
 hydrochloric acid, and allowed to 
 stand in a warm place until the sul- 
 phides and free sulphur have collected 
 and the solution smells but faintly 
 of sulphuretted hydrogen. It is then 
 filtered on the felt, and the precipitate, if it contains 
 
 FIG. 97. 
 
358 COMPLETE ASSAY OF IKON OKES. 
 
 much free sulphur, is treated with carbon disulphide r 
 transferred with the felt to a small beaker, and treated 
 with hydrochloric acid and potassium chlorate, or aqua 
 regia, which dissolves the arsenic and antimony very 
 readily. 
 
 A little tartaric acid is added to keep the antimony 
 in solution, the whole is diluted and filtered, excess of 
 ammonia is added, and then magnesium mixture, and 
 the arsenic precipitated as ammonio-magnesium arseniate 
 with the same precautions that were used in the precipi- 
 tation of phosphoric acid. The precipitate is filtered on 
 the felt, washed with dilute ammonia, dried at 100 C., and 
 weighed as ammonio-magnesium arseniate. 
 
 It is then heated very gradually at first, and finally 
 to a full red, and weighed as magnesium arseniate. The 
 first weight obtained multiplied by 0*3947, and the second 
 by 0-4839, gives arsenic. If the heat is applied carefully 
 and slowly enough at first the latter result is most apt 
 to be -the correct one. The filtrate which contains the 
 antimony is acidulated with hydrochloric acid, and the 
 antimony is precipitated by sulphuretted hydrogen with 
 the usual precautions, filtered on a small disc of paper on 
 the bottom of the perforated crucible, dried, and the pre- 
 cipitate and paper are treated in a porcelain crucible 
 with fuming nitric acid, evaporated to dryness, ignited, and 
 weighed as antimony oxide, which multiplied by 0'7922, 
 gives antimony. 
 
 Titanic Acid. 
 
 As titanic acid is found to interfere with the estima- 
 tion of phosphoric acid, so the latter is found in many 
 cases to absolutely prevent the precipitation of the 
 former. When the ore contains only a small amount of 
 titanic acid 5 grammes are treated with hydrochloric 
 acid, evaporated to dryness, re-dissolved in dilute hydro- 
 chloric acid, filtered, the insoluble residue dried, ignited, 
 and treated in the crucible with hydrofluoric and sulphuric 
 acids. The solution in the crucible is evaporated down 
 until all the hydrofluoric and sulphuric acids are driven 
 
TITANIC ACID. 359 
 
 off, ignited, and the residue is fused with sodium car- 
 bonate. The fused mass is treated with hot water, which 
 dissolves the sodium phosphate, aluminate, and excess of 
 sodium carbonate,* leaving insoluble the sodium titanate 
 and any ferric oxide, calcium carbonate, barium carbonate, 
 &c. The insoluble portion is filtered off, washed, dried, 
 and put aside to be added to the titanic acid in the soluble 
 portion. The filtrate obtained after the first treatment of 
 the ore with hydrochloric acid, which contains the great 
 mass of the iron with the titanic acid which dissolves with 
 it, is deoxidised by ammonium sulphite, the excess of sul- 
 phurous acid driven off, and the ferric phosphate, titanate, 
 and small excess of ferric oxide are precipitated by am- 
 monium acetate exactly as in the estimation of the phos 
 phoric acid. This precipitate is filtered off, washed; 
 dried, and fused with sodium carbonate, the fused mass 
 is treated with water, filtered, and the insoluble portion 
 containing ferric oxide and sodic titanate is washed and 
 dried. The two filters containing the sodic titanate are 
 separated as carefully as possible from the adhering pre- 
 cipitates, ignited in a platinum crucible, the precipitates 
 are added, and the whole is fused at a low heat with 
 15-20 times the weight of acid potassium sulphate. The 
 heat is gradually increased to a dull red, and kept at 
 this point until the fusion was perfectly clear and of a 
 reddish-brown colour. It is then cooled and treated 
 with a large excess of cold water, containing a few drops 
 of sulphuric acid, with constant stirring until nothing re- 
 mains undissolved, except, perhaps, a little silica. It 
 is then filtered, washed with cold water, the filtrate is 
 diluted to about 500 c.c., and boiled in a large platinum 
 dish or beaker with the constant addition of strong aque- 
 ous solution of sulphurous acid, until all the titanic acid 
 is precipitated. The precipitate is allowed to settle, the 
 clear solution is decanted through a filter, the pre- 
 cipitate is boiled up with water, thrown on the filter, 
 and washed as titanic acid. The filtrate is boiled again 
 with the addition of sulphurous acid, and any further pre- 
 
 * Any tungsten in the ore would be in this solution as sodium fcungstate. 
 
360 
 
 COMPLETE ASSAY OF IRON ORES. 
 
 cipitate filtered off and added to the first. It is found 
 .that when the solution is kept at the boiling-point during 
 the filtration, and the precipitate washed with boiling 
 water, there is no tendency to pass the filter ; whereas, 
 if the solution is allowed to get cold, the filtrate is 
 nearly always cloudy. To prevent the precipitation of 
 ferric oxide with the titanic acid, it is necessary to keep 
 an excess of sulphurous acid in the solution ; but if in 
 spite of this precaution the ignited titanic acid is 
 coloured with ferric oxide, it is fused again with potas- 
 sium sulphate and re-precipitated. In the analysis of ores 
 containing much titanic acid it is preferable to fuse the 
 ore at once with sodium carbonate, treat with water, filter, 
 wash and dry the residue, separate and burn the filtrate, 
 fuse the residue and filter ash with potassium sulphate, and 
 estimate the titanic acid as before. 
 
 > 98< 
 
 Estimation of Specific Gravity. 
 
 A great many experiments were made both on lumps 
 and on powdered ore, and it was decided that in commer- 
 
 cial samples good results could 
 be obtained only by using the 
 powdered material. The work 
 was done and the flask herein 
 described was made by Mr. James 
 Hogarth. About 20 grammes 
 of the powdered sample are 
 weighed out and transferred to 
 the specific gravity flask, the con- 
 struction of which is shown in 
 fig. 98. This flask is made 
 with the view of overcoming two 
 difficulties which occur in using 
 the common flask, viz. the ex- 
 pansion and overflow consequent 
 on transferring the flask at 60 
 P. to the higher temperature of 
 
 SPECIFIC. BRAVITV FLASK 
 
 the balance-case, and the necessity for waiting until the 
 
SPECIFIC GRAVITY. 361 
 
 finely divided mineral has settled before the stopper 
 could be inserted without loss of material. These ends 
 are successfully met by melting on a capillary tubulus to 
 the lower part of the neck, and grinding in a stopper 
 having a small bulb above the capillary to allow for 
 expansion. 
 
 Having transferred the weighed quantity of ore to the 
 flask, enough water is added to cover it, and the tem- 
 perature is raised almost to the boiling-point by placing 
 the flask in a water-bath. To insure complete expulsion 
 of air the flask is now placed under a bell-jar, connected 
 with the aspirator, and allowed to boil for a few moments 
 at the reduced pressure. It is then filled up with water 
 almost to the tubulus, cooled, tlie stopper is inserted, 
 and by suction it is filled up slightly above the mark on 
 the capillary part of the stopper. When the thermo- 
 meter is stationary at 60 F. the volume is adjusted by 
 touching the capillary end of the tubulus with blotting- 
 paper, or by bringing a drop of water to it as the case 
 might require. The flask is then dried, transferred to 
 the balance-case, and, after a sufficient time has elapsed 
 to allow it to take the temperature of the room, weighed. 
 
 If W be the weight of ore taken, W the weight of ore 
 4- water at 60, and K the weight of the flask + water at 
 60 then 
 
 Q 
 
 Sp.gr- = = 
 
 To obtain K the flask is nearly filled with boiled water and 
 treated exactly as described above. 
 
 The Perforated Crucible. 
 
 In processes in which the use of filter-paper is delete- 
 rious or inconvenient, nitrations and ignitions have been 
 made with the aid of the asbestos filter devised by Mr. 
 Gooch, which consists essentially of a felt of asbestos 
 deposited upon a perforated surface of platinum.* 
 
 * For a full description of the process see ' Chemical News,' xxxvii. 
 D. 181. 
 
362 
 
 COMPLETE ASSAY OF IRON ORES. 
 
 FIG. 99. PERFORATED CRUCIBLE. 
 
 The mode of preparing the filter is as follows: 
 Asbestos of fine silky flexible fibre is scraped longitudi- 
 nally (not cut) to a 
 fine short down, 
 which is purified by 
 boiling in strong hy- 
 drochloric acid and 
 washing by decanta- 
 tion. A platinum 
 crucible with bottom 
 perforated with nu- 
 merous fine holes is 
 fitted to the upturn- 
 ing end of a piece of 
 soft rubber tubing, 
 the other end of 
 which is stretched 
 over the top of a 
 ' ' f l funnel as shown in 
 
 fig. 99, and the neck of the latter passes through the 
 stopper of a vacuum flask in the usual way. The vacuum 
 pump having been put in action, a little of the prepared 
 asbestos suspended in water is poured into the crucible, 
 and the pump is attached, when the asbestos at once 
 assumes the condition of a firm compact layer, which is 
 washed with ease under the pressure of the pump. 
 
 After washing the felt, the crucible with the felt ad- 
 hering is removed from the funnel, ignited, cooled, and 
 weighed as usual ; then set again in place in the rubber- 
 holder, taking care that the pump is working, and the liquid 
 to be filtered is poured into the crucible. The crucible, 
 felt, and adhering precipitates are, after washing, ignited, 
 or merely dried, as the case demands, with no further care 
 than the nature of the precipitate itself necessitates. 
 
 During the ignition of precipitates it is frequently 
 necessary to prevent the contact of reducing gases with 
 the perforated bottom, and this is accomplished by placing 
 the crucible either upon a platinum crucible cover or 
 within a second crucible, or better by using a form of 
 
CARBON ETC. IN IRON AND STEEL. 363 
 
 crucible which is conical in shape and provided with a 
 cap removable at pleasure, to cover the bottom during 
 ignitions. When a larger filtering surface than the bottom 
 of a crucible affords is needed a cone of platinum, per- 
 forated and fitted to a rubber seat, is substituted for the 
 crucible. 
 
 7. Estimation of Carbon, Sulphur, Silicon, Phosphorus, 
 fyc., in Metallic Iron and Steel. 
 
 Boring and Sampling. Samples of metallic iron for ana- 
 lysis are always obtained by boring, and the proper sam- 
 pling of these borings is as difficult a task as the corresponding 
 one of sampling a heap of ore. As is well known, cast-iron 
 borings are a mixture of small particles of iron with more 
 or less of finely divided graphite separated from the sur- 
 faces of these small particles during the process of boring. 
 The amount of graphite thus mechanically mixed is in 
 all cases quite large enough to cause serious difficulty 
 when the problem is to obtain an average sample of a 
 given lot of borings for carbon estimation. This diffi- 
 culty arises from the necessity of obtaining a uniform 
 mixture of the heavy and comparatively coarse borings 
 with the finely divided and light graphite, and of remov- 
 ing a sample for analysis without disturbing the uniformity 
 of this mixture. 
 
 The difficulties here presented have been investigated 
 by Mr. Porter W. Shinier. It was first sought to over- 
 come the difficulty by having the borings made very fine, 
 so that they might have more nearly the size of the par- 
 ticles of graphite. It was not practicable, however, to 
 secure this uniformity of size, since much of the graphite 
 is in the form of the finest dust. At all events, the dupli- 
 cate estimations of carbon in these fine borings, made 
 with every precaution by combustion in oxygen, frequently 
 showed differences too large to be accounted for in any 
 other way than by imperfect sampling. 
 
 Very coarse borings were also tried, with the idea that 
 the graphite separated from these large and heavy pieces 
 
364 ASSAY OP IRON ORES IN THE WET WAY. 
 
 would perhaps be inappreciable ; but it was found that 
 quite enough was separated to vitiate the results. 
 
 Various methods of mixing were tried, both in the 
 bottle and on glazed paper ; but in all these mixtures it 
 was found that duplicate samples seldom contained the 
 same proportion of the mechanically mixed graphite ; and 
 when they did there could be no certainty that it was the 
 true proportion. All this was shown many times over by the 
 failure of duplicates to agree closely, the difference being 
 sometimes as much as O2 per cent. It was finally clear 
 that nothing was to be hoped from these methods. Even 
 if we could secure a perfect mixture containing in every 
 part its due proportion of graphite, it would run a great 
 risk of being destroyed the moment a spatula was inserted 
 to remove a sample ; for the least agitation causes some 
 of the graphite to fall through between the coarser bor- 
 ings. 
 
 The sampling difficulty was finally overcome in this 
 way : The borings were carefully poured out into a large 
 porcelain crucible or dish, and enough alcohol is added to 
 merely moisten them (2 c.c. alcohol to 450 grains). Mix 
 thoroughly for about five minutes. Bernove with a spatula 
 as many samples of the moist borings as are needed, weigh- 
 ing them out roughly so as to get approximately the weights 
 desired. Dry off the alcohol and weigh accurately* The 
 samples thus obtained perfectly represent the average of 
 the original sample, for the alcohol simply serves tempo- 
 rarily to hold the graphite where it belongs, upon the 
 surface of the borings. 
 
 As the alcohol evaporates from the original sample 
 the graphite again falls away from the borings, and the 
 moistening must be repeated when new samples are to be 
 removed. 
 
 The following total carbon results show the im- 
 provement : 
 
 i. ii. 
 
 Mixed dry in the bottle . > . > . 3'68 3'84 p.c. 
 
 Moistened with alcohol and mixed . . 3'97 3-99 
 
 The first results are an extreme case of unsatisfactory 
 duplicates ; and they also show the tendency of the gra- 
 
CARBON, SULPHUR, SILICON, PHOSPHORUS, ETC. 365 
 
 phite to go to the bottom of the bottle when mixed 
 dry. The mechanically mixed graphite, approximately 
 estimated by removing the iron by means of a magnet from 
 samples taken moist, was 0-76 and 0-79 per cent, in dupli- 
 cate estimations. 
 
 After shaking up the dry borings a few times, and re- 
 moving nearly all from the bottle, the remaining sample 
 contained 1*29 per cent, mechanically mixed graphite. 
 
 The other estimations, such as phosphorus and silicon 
 (iron excepted), are perhaps not appreciably affected 
 when samples are taken dry. But whenever samples 
 are thus taken, the borings which remain are for ever 
 vitiated for carbon estimations which shall truthfully 
 represent the original sample. In very careful work, 
 therefore, and especially when there is any prospect of 
 subsequent carbon estimations to be made, it is neces- 
 sary to moisten the borings when weighing out samples 
 for phosphorus and silicon, or to have two sets of borings 
 made, one of which is to be reserved for carbon esti- 
 mations. For iron estimations the samples should always 
 be taken moist. 
 
 It was suggested by Mr. Frank Firmstone, Superin- 
 tendent of the Glendon Iron Works, to determine whether 
 or not there is any appreciable loss of graphitic dust while 
 making cast-iron turnings according to the method devised 
 by him, in which the turnings are taken equally from the 
 whole section of a piece of pig iron, the outside being first 
 removed. Through the kindness of Mr. Firmstone, Mr. 
 Shinier was enabled to procure from the same piece of 
 pig iron a sample of turnings taken dry, a second sample 
 kept moist with alcohol to prevent loss of dust, and several 
 sections -^ inch in thickness. 
 
 The following are the total carbon results from these 
 samples : 
 
 I. Turnings taken dry, but sampled moist. 
 
 II. Turnings taken moist. 
 
 III. Thin section. 
 
 i. ii. in. 
 
 4-132 4-122 4-031 
 
 4-123 4-123 4-038 
 
 4-111 4-103 4-108 
 
3G6 ASSAY OF IRON ORES IN THE WET WAY. 
 
 The results under I. and II. show that, with care in 
 catching the turnings, and avoidance of draughts of air, 
 there is no appreciable loss of graphitic dust. The first 
 two results on the thin section are low, because of in- 
 complete decomposition of the iron, when, after forty 
 hours, hydrochloric acid was added to complete the solu- 
 tion. The third result is about right ; but complete decom- 
 position was obtained in this case by frequent and long- 
 continued stirring. While, therefore, it is possible to 
 obtain the correct total carbon by use of a thin section, it 
 offers no advantage in point of time, since borings may 
 be dissolved in neutral and cold copper and ammonium 
 chloride solution in fifteen to twenty minutes by constant 
 stirring. 
 
 Some chemists, with a keen appreciation of the sam- 
 pling difficulty, have proposed to overcome it by using 
 clippings of the iron from -J to y 1 ^ inch in thickness. 
 While it is no doubt possible, in most cases, to obtain the 
 true result by this means, it is open to the objection of 
 being very slow when these clippings are to be dissolved 
 in neutral copper and ammonium chloride solution. Fur- 
 thermore, the weight of iron taken for carbon estima- 
 tion in cast iron being limited by convenience to about 
 50 grains, a very few clippings make up this weight, 
 and these represent only those parts of the sample from 
 which they happen to come. While there may not be an 
 appreciable difference in the composition or in the condi- 
 tion of the carbon of the different parts of a section of pig 
 iron, it is yet possible that there may be, and in this case 
 it is easy to be on the safe side. 
 
 When turnings are taken equally from the whole sec- 
 tion of a piece of pig iron, and are then moistened with 
 alcohol and thoroughly mixed, every part of the section 
 has its due representation in a sample of 50 grains. 
 
 It may not be out of place to add that, in sampling ores, 
 and, in fact, any similar mixture of particles of different 
 size, composition, and specific gravity, it is advisable to 
 moisten with water before mixing. This secures a uniform 
 mixture of coarse and fine parts from which a true sample 
 
ESTIMATION OF THE TOTAL CARBON. 367 
 
 can easily be removed, prevents loss of dust, and is in every 
 way more satisfactory than mixing dry. 
 
 In pulverised ores it is very common for the finest part 
 to consist principally of lighter gangue material ; and the 
 problem of securing a true sample of such a mixture is 
 similar to that of obtaining a true sample of a mixture of 
 cast-iron borings and graphite. 
 
 Carbon. This is a problem of considerable difficulty, 
 and to secure accurate results many special precautions are 
 necessary, owing to the large preponderance of the iron 
 over the carbon present. The carbon may be present in 
 two forms as combined carbon, and as free or graphitic 
 carbon. The estimation may be of the total carbon 
 present, or of either the combined or graphitic separately, 
 and the method of analysis adopted will have to be selected 
 accordingly. The following is a description of the most 
 satisfactory processes which have been devised for these 
 estimations. 
 
 A. Estimation of the Total Carbon. Mr. Addison B. 
 Clemence employs the following method for estimating 
 carbon in steels. The cut (fig. 100) shows the form of 
 apparatus adopted. It is made of platinum. 
 The following is the process : 
 
 Dissolve from 50 to 80 grains of the borings in double 
 chloride of copper and ammonium, using 560 grains of 
 the salt to 40 ounces of water for 46 grains of steel. After 
 the separated copper is completely dissolved, filter on to 
 a plug of asbestos placed at 5, and wash thoroughly with 
 hot water. Any carbon that adheres to the sides of the 
 tube may be swept down by moistened asbestos. The tube 
 is then placed in an air-bath, and dried at a temperature 
 of from 350 to 380 F. for about one hour. A hard 
 rubber cork through which is passed a glass tube is in- 
 serted at e, the oxygen gas passing from c to a. Around 
 the tube at c is a single thickness of filter-paper, about 
 two inches wide, kept wet by a stream of water supplied 
 by a reservoir on a shelf above. Heat is applied at b for 
 one half-hour, at the end of which time the potash bulb 
 is ready for the balance. One Bunsen burner is sufficient 
 
368 
 
 ASSAY OF IKON ORES IN THE WET WAY. 
 
 for the combustion. The same precautions are taken 
 to dry the gas before entering the platinum tube, as 
 well as before entering the potash bulb, as in the case 
 with the porcelain tube method. 
 
 ; Six burners of a combustion furnace 
 will consume 14 ft. of gas in one half- 
 hour, while one Bunsen burner will con- 
 sume 2 ft. in the same time. 
 
 The following table shows some re- 
 sults obtained by the " platinum tube " 
 process : 
 
 A is a Swedish Bessemer iron contain- 
 ing 0*10 per cent, of carbon ; 
 
 B, an American Bessemer, with 0-18 
 per cent. ; and 
 
 C, an American Bessemer, with 0'50 
 per cent. ; 
 
 all obtained by the "porcelain tube" 
 process. 
 
 <----* 
 
 A. 
 
 0-09 
 0-10 
 0-10 
 0-10 
 
 B. 
 
 0-18 
 0-18 
 0-16 
 0-17 
 
 c. 
 
 0-49 
 0-51 
 0-49 
 0-44 
 
 Weyl's very ingenious method for the 
 estimation of the total carbon is founded 
 upon the fact that a piece of iron at- 
 tached to the positive pole of a galvanic 
 battery, and suspended in hydrochloric 
 acid, is dissolved, while the hydrogen is 
 given off at the negative pole. The for- 
 mation of hydrocarbons, and a consequent 
 loss, is in this manner prevented. One 
 great advantage in this method is that 
 the iron does not require to be in pow- 
 der. A piece of iron 2 to 4 grammes in 
 weight, attached to the positive pole of 
 ** a Bunsen's cell, is suspended in dilute 
 
 hydrochloric acid, just below the surface of the liquid. 
 
 From the negative pole hydrogen passes off, while the 
 
ESTIMATION OF GRAPHITE. 369 
 
 iron dissolves quite quietly, and the strong solution of 
 ferrous chloride formed may be seen falling in a regular 
 stream through the lighter liquid. The iron is dissolved 
 in about twenty-four hours, and the carbon is left behind 
 in the same shape as the piece of metal from which it 
 was derived. In Weyl's earlier experiments it was found 
 that some of the liberated carbon at the positive pole 
 was carried over to the negative pole by the mechanical 
 working -of the stream. To prevent this, a diaphragm 
 of bladder or parchment paper is interposed between 
 the two, which entirely obviates the possibility of loss 
 in this way. 
 
 B. Estimation of the Graphite. Dr. Eggertz reduces 
 1 gramme of iron to small pieces, mixes with 5 grammes 
 of pure iodine and 5 c.c. of water in a small flask, covered 
 with a watch-glass, and placed in ice-cold water before 
 adding the iron. It is to be kept for twenty-four hours at 
 C., and frequently stirred meanwhile. By keeping the 
 liquid cold, no carburetted hydrogen is produced. The 
 greater the amount of silicon in the iron the greater is 
 the tendency to the production of carburetted hydrogen. 
 The residue of carbon and silica left after the iron is dis- 
 solved is collected on a filter of known weight, when it is 
 dried at from 95 to 100 C., and washed thoroughly with 
 hot water. After twelve hours, it is to be washed with 
 a mixture of hydrochloric acid and twice its volume of 
 water, heated to 70 or 80 C., until the filtrate ceases to 
 give a blue colour with ferrocyanide solution. The object 
 of leaving the residue for twelve hours is to allow any 
 small particles of iron remaining undissolved by the iodine 
 being oxidised by atmospheric air, and prevent disengage- 
 ment of hydrogen when the hydrochloric acid is added. 
 After the hydrochloric acid is washed out of the filter, it 
 is dried, with its contents, at 95 to 100 C., until constant 
 in weight. This weighing gives the amount of the carbo- 
 naceous residue and silica (but not the whole of the silica, 
 because some part of it would have been dissolved), and 
 by burning the carbon away and weighing the silica the 
 weight of the carbonaceous residue may be ascertained. 
 
 B B 
 
370 ASSAY OF IRON ORES IN THE WET WAY. 
 
 The carbonaceous residue may consist either of graphite 
 or of the compound of carbon, iodine, and water already 
 mentioned, if the carbon was combined with the iron. To 
 ascertain which is the case, 1 gramme of iron is dissolved in 
 15 c.c. of hydrochloric acid (1*12 density) in a flask covered 
 with a watch-glass, and, when the iron is dissolved, the solu- 
 tion is boiled for half an hour. All the carbon combined with 
 the iron is disengaged in the form of carburetted hydrogen 
 gas, while the graphite and silica remain. If the carbona- 
 ceous residue left after dissolving the iron comes in con- 
 tact with atmospheric air before the liquid is boiled, it is so 
 altered that it is not dissolved and disengaged as gas. The 
 graphite that remains after boiling the liquid is collected on 
 a filter of known weight, washed, dried, and weighed. It is 
 then burnt, and the residual silica weighed to ascertain the 
 quantity of graphite. Very satisfactory results have been 
 obtained by this method. The differences do not amount 
 to more than Ol per cent. When the quantity of carbon 
 to be estimated is very small, more than 1 gramme of iron 
 must be used in the analysis. 
 
 Mr. Tosh gives the following process for the estimation 
 of graphite. Two to three grammes of iron are treated 
 with dilute hydrochloric acid, and when the solution 
 approaches completion a considerable quantity of strong 
 acid is added to separate the last portions of iron and 
 manganese. The insoluble matter, consisting mostly of 
 graphite, is collected on a carefully weighed filter, washed 
 with hot water, dilute hydrochloric acid, solution of caustic 
 soda, and hot water again, successively, and lastly with 
 alcohol and ether to remove oily hydrocarbons. (By wash- 
 ing with dilute acid and with alkali, the iron and silica or 
 oxide of silicon are separated.) After drying at 120 C., 
 the filter and graphite are weighed, and burned away. 
 The small residue (a mere trace of silica or titanic acid) is 
 weighed, and this weight subtracted from the first gives 
 the amount of graphite. The results obtained agree very 
 closely. 
 
 In washing the graphite with solution of soda, there is 
 always a brisk effervescence, due to the oxidation of oxide 
 
ESTIMATION OF COMBINED CARBON. 371 
 
 of silicon to silicic acid, by decomposition of water, with 
 consequent liberation of hydrogen. 
 
 C. Estimation of Combined Carbon. When steel, or 
 pig-iron containing carbon in chemical combination, is dis- 
 solved in nitric acid, a soluble brown colouring matter is 
 formed, whose colouring power is very intense, and the 
 solution assumes a tint which is dark in proportion to the 
 quantity of the chemically combined carbon. Iron and 
 graphite (or free carbon) do not influence this colouration ; 
 for the solution of nitrate of iron is only slightly greenish, 
 unless extremely concentrated, and graphite is insoluble in 
 nitric acid. 
 
 Thus, in dissolving two pieces of different steels of the 
 same weight in nitric acid, taking care to dilute the darker 
 solution until the two liquids present exactly the same 
 colour, it is very evident that the more highly carburetted 
 steel will furnish the larger quantity of liquid, and that the 
 proportion of the volumes will indicate the relative pro- 
 portion of colour in the two steels. If now the composi- 
 tion and content of carbon of one of the steels is known, 
 the absolute percentage of carbon in the other steel may 
 be immediately deduced. Dr. Eggertz has applied these 
 reactions to a method of estimating the combined carbon. 
 To obtain trustworthy results certain precautions must be 
 taken. 
 
 In a cylindrical test-tube, dissolve gradually in the 
 cold 10 centigrammes of wrought-iron, steel, or cast iron, 
 reduced to a fine powder, in 1-J to 5 c.c. of nitric acid of 
 1*2 sp. gr. The use of nitric acid containing hydrochloric 
 acid must be avoided, because the solution of iron would 
 have a yellow tint. In proportion as the metal contains more 
 carbon, more nitric acid must be used. After some time, 
 when the chief part of the metal appears to be attacked, 
 place the tube in a water bath, and warm it to 80 C., in 
 such a position that only the lower part of the tube is in 
 contact with the warm water ; a movement takes place in 
 the acid, which favours its reaction upon the metal; a 
 slight disengagement of carbonic acid from all the particles 
 of carbon may be observed. The operation should always 
 
 B 2 
 
372 EGGERTZ'S COLOUR TEST FOR THE 
 
 be conducted under the same circumstances as to heat and 
 length of time. The evolution of gas having ceased (in 
 operating upon steel, the reaction must continue two or 
 three hours), place the tube in a large vessel filled with 
 water, to bring the solution always to the same temperature. 
 This precaution is indispensable, because the same liquid is 
 darker when warm than when cold. Afterwards, pour off, 
 as exactly as possible, the clear liquid into a graduated 
 burette. Upon the black residue remaining in the tube 
 pour some drops of nitric acid, and heat carefully over a 
 lamp. If there is no further liberation of gas, the residue 
 consists of nothing but graphite or silica. Cool the new 
 solution, and add it to that which is already in the 
 burette. 
 
 The liquid is then diluted with water until its colour 
 corresponds exactly with that of the normal liquid, which 
 latter should be of such a degree of concentration that 
 each c.c. represents O'OOOl grm. of carbon. If, for instance, 
 this normal liquid is prepared from cast steel containing 
 exactly 0*85 per cent, of carbon, 1 decigramme of that 
 steel must be dissolved in 8*5 c.c. of nitric acid ; lOOgrms. 
 of steel containing 85 centigrammes of carbon would thus 
 be dissolved in 8500 c.c. of the normal solution, 100 c.c. 
 of that solution would represent 1 centigramme of carbon, 
 and, consequently, 1 c.c. of the normal solution would 
 represent O'OOOl grm. of carbon. To compare the normal 
 solution with the solution of iron under examination, it 
 should be contained in a tube of the same kind, and when 
 the two tubes are held together by daylight before a thin 
 sheet of paper, the colour should be exactly the same in 
 both of them. For the preparation of normal colour solu- 
 tions see pp. 382-388. 
 
 The Eggertz colour test for estimating carbon in steel 
 is perhaps used more extensively commercially than any 
 other analytical process. This is particularly the case in 
 Bessemer steel works, where every blow is tested to ascer- 
 tain its percentage of carbon. The number of estimations 
 thus amounts even in a small works to 25 daily. 
 
 In carrying out the analytical process, which is too 
 
ESTIMATION OF CARBON IN STEEL. 
 
 373 
 
 FIG. 101. 
 
 well known to require description, it is of course necessary 
 to preserve the identity of each blow-number ; and this is 
 effected by dissolving the steels in a square shallow copper 
 water-bath provided with a lid perforated with holes in 
 which the tubes are inserted in a standing position, the 
 first hole indicating the starting-point of the day's blows. 
 
 This form of bath is objectionable from several points 
 of view. 
 
 Mr. J. Oliver Arnold has consequently designed a 
 bath which two years' constant use has proved to have 
 none of the disadvantages 
 connected with the old form 
 of bath. 
 
 The bath consists of a 
 cylinder of glass 5 inches high 
 by 5 inches in diameter, closed 
 at one end to contain the 
 water ; this is heated on an 
 iron pi ate or sand-bath nearly 
 to boiling-point (90-95). 
 On the top is a perforated 
 cover of glazed white earth- 
 enware ; this cover contains 
 17 holes, numbered 1 to 15 
 in burnt-in black figures, 
 and the other two holes are 
 marked S for the standard 
 steels. 
 
 The tubes containing the 
 steels and nitric acid are suspended in the water to the 
 proper depth by slipping on the tube a rather tightly 
 fitting cylinder of india-rubber f inch long ; this india-rubber 
 serves the double purpose of keeping the ends of the tubes 
 off the bottom of the bath, bumping thus being avoided, 
 and of preventing the bobbing up of the tubes by flota- 
 tion (see figure). 
 
 The apparatus, being of glass and glazed earthenware, 
 can be washed perfectly clean every day. The tubes are 
 suspended vertically in the water to any desired depth. 
 
374 ESTIMATION OF CAEBON IN STEEL. 
 
 The clear numbering enables the selection of any given 
 blow about which 4 a hastener ' has been received with- 
 out chance of error, and the progress of the solution of 
 the steels in the tubes can be watched without removing 
 them. 
 
 Except from mechanical breakage the apparatus is 
 practically indestructible. In cases where the daily car- 
 bons exceed 15, a second and third bath may be employed, 
 the numbering following on, or one large bath could be 
 made ; this, however, might prove inconveniently large. 
 
 In cases where only a few odd carbons are required a 
 smaller bath perforated for 6 steels and a standard would 
 suffice. 
 
 Several modifications of Eggertz's method have been 
 proposed. The most successful of them, affording exceed- 
 ingly accurate results, was communicated to the ' Journal 
 of the Franklin Institute ' for May 1870 by J. Blodget 
 Britton, and has been tested for a considerable time. In- 
 stead of a single tube, containing a standard solution for 
 comparison, as suggested by Eggertz, a number of tubes- 
 
 FIG. 102. 
 
 A 
 
 8aJJjJLIlLttl 
 
 having their solutions differently standardised, one from 
 the other, are employed. These are arranged securely in 
 a wooden frame, with spaces between for placing the tube 
 containing the solution to be tested, and forming together 
 a convenient portable instrument called a colorimeter a 
 representation of such an instrument being shown in the 
 annexed cut. The position of the tube containing the 
 solution to be tested is shown at A. The tubes are fths 
 of an inch in diameter, and 3-| inches in length, filled with 
 water and alcohol coloured with roasted coffee, and herme- 
 tically sealed. The solution in the tube to be left has its 
 colour to correspond exactly with one produced by 1 
 
BRITTOIs T S METHOD. 375 
 
 gramme of iron, containing 0'02 per cent, of combined 
 carbon, dissolved in 15 c.c. of nitric acid. The solution 
 in the tube next to it has its colour to correspond with 
 one produced by the same quantity of iron, but containing 
 0-04 per cent, of combined carbon ; and so with each of 
 the other tubes, increasing 0*02 per cent, of carbon in 
 regular succession to the right, the last reaching 0*3 per 
 cent., as indicated by the figures on the upper rail of the 
 instrument. On the back of the instrument, and for 
 the purpose of partially screening the light and allowing 
 the different shades of colour to be distinctly discerned, 
 there is tightly stretched between the rails some fine white 
 parchment paper. This screen is not shown by the cut, 
 but it serves a very important purpose. The process is 
 conducted as follows : 1 gramme of the finely divided 
 metal is put into a tube of about 1-J inch in diameter 
 and 10 inches long, and digested for fifteen or twenty 
 minutes in 10 c.c. of nitric acid of a little more than 1-20 
 specific gravity, free from chlorine. The solution is then 
 cautiously poured into a beaker, and a small portion of 
 metal which remains undissolved and adheres to the bottom 
 of the tube is treated with 5 c.c. of fresh acid, exposed to 
 a gentle heat till completely dissolved, and added to the 
 other. The contents of the beaker, when sufficiently cool, 
 are filtered through two thicknesses of German paper (not 
 previously moistened, and of a diameter not exceeding 4^ 
 inches) into a tube about 5 inches long and of precisely 
 the same diameter as those in the instrument. After the 
 filtered solution has remained for some minutes at the tem- 
 perature of the atmosphere, and its colour become fixed, 
 the tube is placed in the instrument and the carbon esti- 
 mated by a comparison of shades : the estimation may 
 be made readily as close as O'Ol per cent. Heat should 
 not be applied in the first instance to facilitate the solution 
 of the metal, because a high temperature is apt to cause a 
 slight loss of colour. Two thicknesses of paper are taken, 
 because one alone is liable to break ; and the paper should 
 be used dry, for, if previously wetted, the water will 
 weaken the colour of the solution ; and it ought to be cut 
 
376 ESTIMATION OF CARBON IN STEEL. 
 
 to a size not exceeding 4-^ inches, to prevent undue absorp- 
 tion. If the metal to be examined contains more than 
 0-3 per cent, of carbon, O5 gramme, or less of it, may be 
 taken, or the solution may be diluted with an equal volume 
 or more of water and the proper allowance made ; or an 
 instrument of higher range may be used. On the other 
 hand, if the metal contains a very small percentage of 
 carbon, 2 grammes of it may be taken. For preparing the 
 standard solutions (the normal ones begin to lose colour 
 after some hours), caramel dissolved in equal parts of 
 water and alcohol, as suggested by Eggertz, answers well ; 
 but with roasted coffee as the colouring matter the true 
 shades may be obtained. (See also pp. 382-88.) 
 
 An important paper by Mr. J. E. Stead was read in 
 1883 before the Iron and Steel Institute on a new method 
 of estimating minute quantities of carbon in iron and steel, 
 and on a new form of chromometer. 
 
 The following a,re extracts from Mr. Stead's paper : 
 
 1. ESTIMATION OF MINUTE QUANTITIES OF CARBON. As is 
 well known, it is impossible to estimate with accuracy 
 minute quantities of carbon by the ordinary colour method, 
 owing to the colour of the nitrate of iron present, which 
 interferes so as to make it impossible to judge of the colour 
 due to carbon. 
 
 Having been engaged in some careful investigations on 
 the nature of the colouring matter which is produced by 
 the action of dilute nitric acid upon white iron and steel, 
 it was found it had the property of being soluble in 
 potash and soda solutions, and that the alkaline solution 
 had about two and a half times the depth of colour pos- 
 sessed by the acid solution. This being so, it was clear 
 that the colouring matter might readily be separated from 
 the iron, and be obtained in an alkaline solution, by 
 simply adding an excess of sodium hydrate to the nitric 
 acid solution of iron, and that the colour solution thus 
 obtained might be used as a means of estimating the 
 amount of carbon present. Upon trial this was found to 
 be the case, and that as small a quantity as 0-03 per cent, 
 carbon could be readily estimated. 
 
STEzVDS METHOD. 077 
 
 The method as now in use is conducted as follows : 
 Standard solution of nitric acid, T20 sp. gr. 
 Standard solution of sodium hydrate, 1*27 sp. gr. 
 
 One gramme of the steel or iron to be tested is weighed 
 off and placed in a 200 c.c. beaker, and, after covering 
 with a watch-glass, 12 c.c. of standard nitric acid are 
 added. The beaker and contents are then placed on a 
 warm plate, heated to about 90 or 100 C., and then 
 allowed to remain until dissolved, which does not usually 
 take more than ten minutes. At the same time a stan- 
 dard iron containing a known quantity of carbon is treated 
 in exactly the same way, and when both are dissolved 30 
 c.c. of hot water are added to each, and 13 c.c. soda 
 solution. 
 
 The contents are now to be well shaken, and then 
 poured into a glass measuring-jar and diluted till they 
 occupy a bulk of 60 c.c. After again well mixing and 
 allowing to stand for ten minutes in a warm place, they 
 are filtered through dry filters, and the filtrates, only a 
 portion of which is used, are compared. This may be 
 done by pouring the two liquids into two separate measur- 
 ing-tubes in such quantity or proportion that upon looking 
 down the tubes the colours appear to be equal. 
 
 Thus if 50 m.m. of the standard solution is poured 
 into one tube, and if the steel to be tested contains, say, 
 half as much as the standard, there will be 100 m.m. of 
 its colour solution required to give the same tint. The 
 carbon is therefore inversely proportional to the bulk 
 compared with the standard, and in the above-assumed 
 case, if the standard steel contained 0-05 percent, carbon, 
 the following simple equation would give the carbon in 
 the sample tested : 
 
 Q-" = 0-025 per cent. 
 100 
 
 Experiments were made upon a steel which contained 
 0-11 per cent, carbon, to ascertain what the influence 
 would be of heating the nitric acid solution for an increas- 
 ing length of time after dissolving on the bath, and it 
 
378 ESTIMATION OF CARBON IN STEEL. 
 
 was found that the carbon colour was not materially 
 affected by heating the acid solution twice as long as was 
 necessary for completely dissolving the carbon compound, 
 and that although the iron is dissolved in five minutes 
 it is evident that some of the carbon compound at first 
 formed escapes solution in that period. 
 
 The next point was to ascertain what effect the use of 
 an excess of nitric acid in dissolving the steel w^ould have 
 on the colouring matter. It was seen that 6 c.c. acid in 
 excess does not materially affect the estimation, but when 
 this is exceeded the colour is reduced in quantity. 
 
 It now became important to know if a greater or less 
 quantity of soda solution would have a different solvent 
 power on the colouring matter. 
 
 To ascertain this, four separate portions of the soft 
 steel were treated alike in dissolving, but to the solutions 
 different quantities of soda solution were added. It was 
 here seen that, as before stated, 13 c.c. sodium hydrate solu- 
 tion is capable of effecting solution of the colouring matter. 
 By using a less amount, however, by experiment it was 
 found that the colour is precipitated with the iron oxide. 
 
 It is very well known that, in the old colour 
 method, very slight traces of hydrochloric acid, if present, 
 alter the character of the colour to such an extent as 
 to make the colour estimation unreliable. It therefore 
 was of interest to ascertain if the same would occur in 
 the alkaline method. Four portions of steel were treated 
 as usual, excepting that to one portion a single drop of 
 hydrochloric acid was added when being dissolved, to a 
 second five drops, and to a third ten drops, but to the 
 last portion no hydrochloric acid was added. 
 
 Th.e following are the results obtained, viz. : 
 
 Hydrochloric Acid. 
 
 1 drop 5 drops 10 drops None 
 
 Carbon . 0-105 p. c. 0-090 p. c. 0-078 p. c. 0-110 p. c. 
 Second Test. 
 
 1 drop 5 drops 10 drops None 
 
 Carbon . 0-356 p. c. 0*338 p. c. 0-324 p. c. 0-410 p. c. 
 
 The colour in each case, and even in that in which the 
 larger quantity of hydrochloric acid was added, was the 
 
STEAD S METHOD. 
 
 379 
 
 same in quality although differing in quantity, showing 
 (1) that the presence of chlorides is harmless, and (2) 
 that nitro-hydrochloric acid, even in small quantities, pre- 
 vents the formation of the full amount of colouring matter 
 capable of being produced by nitric acid alone. 
 
 A large number of samples of low carbon iron have 
 been examined by the alkaline method, including samples 
 of iron taken from the Bessemer converter at the end of 
 the blow before any addition of spiegeleisen. The results 
 are given here 
 
 Blown Iron taken from the Bessemer Converter. 
 No. 1 0-040 p. c. carbon. 
 
 No. 2 
 
 No. 3 
 
 No. 4 
 
 No. 5 
 
 No. 6 
 
 Average 
 
 Average by combustion 
 
 Standard soft steel . 
 Pure iron wire 
 Cleveland iron ship plates 
 Ditto ditto 
 
 The colour solutions from these low carbon irons are 
 different in tint from those obtained from the higher car- 
 bon steels ; and it is important that a low carbon iron be 
 used as a standard for comparison. 
 
 When high carbon steel is heated to redness and chilled, 
 it is well known that the colour from the chilled steel is 
 very much less in quantity than that from the same steel 
 before hardening. The difference, however, is not nearly 
 so marked when there is little carbon present in the steel, 
 as was proved by the following results, viz. : 
 
 Several samples of iron and steel after being drilled 
 were heated to redness and chilled in water, the results 
 before and after being as follows : 
 
 0-036 
 
 
 0-045 
 
 
 0-039 
 
 
 0-061 
 
 
 0-048 
 
 
 0-045 
 
 
 0-048 
 
 
 New Colour 
 
 Method Combustion 
 
 0-120 p. c. 
 
 carbon. 0*122 p. c. 
 
 0-038 
 
 , 
 
 0-055 
 
 > 
 
 0-035 
 
 
 
 Soft steel = soft 0-168 p. 
 
 chilled in cold water 0-158 
 
 chiUed in hot water . 0-168 
 
 Staffordshire square iron bar . O'llO , 
 
 chilled 0-100 , 
 
 flat . 0-069 . 
 
 chilled 0-069 , 
 
 Soft steel 0-077 : 
 
 chilled. . . 0-071 , 
 
 Difference 
 
 >0-010 p. c. 
 
 0-010 
 
 None. 
 
 0-006 
 
380 ESTIMATION OF CARBON IN STEEL. 
 
 It is not often that soft iron or steel is chilled before 
 being placed in the hands of the analyst, but it is satisfac- 
 tory to know that even if they were the results by colour 
 would not be rendered useless. 
 
 When using the new method Mr. Stead found that 
 some steels give a much yellower colour than others, and 
 in course of investigation he discovered that there are 
 present in all nitric acid steel solutions two distinct colour- 
 ing matters, which have been separated arid obtained in a 
 nearly pure state, one of which is bright yellow, resembling 
 potassium chromate, the other being of a dark brown-red 
 colour. In some steel solutions the yellow colour pre- 
 ponderates, and in others the brown. 
 
 2. A NEW FORM OP CHROMOMETER. In comparing 
 colour solutions there are two methods of procedure. 
 The first is that generally adopted in making estima- 
 tions of the carbon by the acid colour process, in which 
 the darker solution is diluted with water until the 
 colours of the two solutions are equal in density that is 
 to say, until the colour is equal per cubic centimetre. 
 The diluted volume is then noted, and the amount of 
 carbon read off in 0*1 per cent. c.c. In the alkaline 
 method it is better to use the second plan, already 
 described, of comparing directly the relative density of 
 the colour solution without dilution, and ascertaining 
 the lengths of the two columns of liquids, which, when 
 examined from the surface, give the same depth of colour. 
 The carbon in this process is, as compared with the stan- 
 dard, inversely proportional to the length of the liquid 
 column. 
 
 If a fixed length of liquid column be used of the solu- 
 tions of carbon and a variable standard column, then, by 
 using a suitable standard solution, the carbon may be de- 
 duced from the length of the latter required to make a 
 colour column equal in depth to the former and the per- 
 centage read directly from a graduated scale. The instru- 
 ment arranged by Mr. Stead is made on this principle; it is 
 extremely simple and easily constructed. It consists of 
 two parallel tubes, which may be of any suitable diameter, 
 
STEAD'S CHROMOMETER. 381 
 
 one of which is contracted at a point 9 inches from the 
 top, and is open at both ends. The lower end passes 
 through an india-rubber cork to the bottom of a 4-oz. 
 bottle, which contains the standard colour solution. A 
 second tube of smaller diameter also passes through the 
 cork into the bottle, the outside end of which is in com- 
 munication with a large syringe. 
 
 Just above the contracted part of the first-mentioned 
 tube a small glazed cylinder of china clay rests. By 
 pressing the syringe the liquid can be forced from the 
 bottle below up this tube. The second tube is about 9 
 inches long, and is closed at the lower end. At this end a 
 small glazed clay cylinder is also placed. When this tube 
 is placed parallel to the first, the length from the open 
 upper ends to the flat surfaces of the clay prisms is equal 
 in each. A small looking-glass at an angle of 45 is fixed 
 above the open ends of the tubes, and the standard tube 
 is graduated into O'Ol part to O15 part. 
 
 The method of working with the apparatus is very 
 simple. The colour solution to be compared is placed in 
 the second tube, with which it is filled up to a certain 
 fixed mark. 
 
 It is only now necessary to squeeze the syringe and 
 force the liquid up the first tube until the colours in the 
 two columns are equal, as can be seen by looking into the 
 mirror above. The height of the standard solution is read 
 off on the graduated scale, which will be the percentage 
 of carbon in the steel or iron under examination. 
 
 The preparation of Inorganic Standards for the 
 Colorimetric Carbon Test. 
 
 Wherever the amount of work renders it practicable 
 the plan of using permanent standard solutions in con- 
 nection with the colorimetric carbon test affords such 
 manifold advantages that it is to be strongly recommended. 
 That it has not attained a wider and more general applica- 
 tion is mainly due to the fact that, as the method is 
 generally employed, the difficulty attending the produc- 
 
382 ESTIMATION OF CARBON IN STEEL. 
 
 tion of the colours, and their doubtful stability when 
 produced, have been only too evident to those who have 
 undertaken their composition. The old method of using 
 standard solutions made from sugar, coffee, &c., has 
 always been hampered by the tendency of the colours to 
 change upon exposure to the light, and the consequent 
 liability to serious error unless they are closely watched. 
 Appreciating these difficulties while recognising the de- 
 sirability of permanent colours, Mr. Magnus Troilius, at 
 the suggestion of Professor F. L. Erckmann, investigated 
 the properties of cobalt, copper, and ferric chlorides with 
 the view of determining their efficacy in the production 
 of standard solutions. His results seem to have been 
 satisfactory, but his treatment failed to eliminate many 
 of the objections that have rendered the use of organic 
 colours undesirable. The following mode of procedure 
 in the making-up of inorganic standards has been found 
 by Mr. W. Eobinson, of Joliet, Illinois, to be most effi- 
 cient in point of accuracy, ease, and celerity of production 
 and stability. 
 
 DAY STANDARD COLOURS. 
 
 Mr. Eobinson uses as a range in illustration the car- 
 bon-contents included between the extremes 0*5 per 
 cent, arid 0-3 per cent, as representatives of Bessemer 
 rail steel that branch of the industry in which the uni- 
 formity and continuity of the product renders the use of 
 permanent standard solutions specially advisable. The 
 principles formulated below will evidently apply to either a 
 higher or lower carbon range. The salts employed are the 
 same as those used by Mr. Troilius, viz. the neutral chlorides 
 of cobalt, copper, and iron. Solutions of these are made 
 as follows : 
 
 Cobalt Chloride. Dissolve in water containing 1 per 
 cent, of free hydrochloric acid (1*12 specific gravity) in 
 the ratio of 1 gramme salt to 1 c.c. water. 
 
 Copper Chloride. Dissolve in water containing 1 per 
 cent, hydrochloric acid (1-12 specific gravity) in the ratio 
 of 1 gramme of the salt to 1^ c.c. water. 
 
PREPARATION OF STANDARD COLOURS. 383 
 
 Ferric Chloride. Dissolve in water containing 2 per 
 cent, hydrochloric acid (1-12 specific gravity) in the ratio 
 of 1 gramme salt to 1 c.c. water. 
 
 The solution in all cases is best effected by the appli- 
 cation of a gentle heat ; and filtering through a ribbed 
 filter assures freedom from turbidity. Free acid is used 
 to prevent the otherwise enhanced liability to decomposi- 
 tion of the chlorides. Should any difficulty be encoun- 
 tered in obtaining a clear solution of the ferric chloride, 
 subsidence of the liquid for a few hours, and then decan- 
 tation, will obviate the trouble. 
 
 Mr. Eobinson proceeds : ' Our purpose is now to 
 make up a set of colours that shall correspond to those 
 made by the nitric acid solution of steels containing in car- 
 bon every alternate point between the extremes of 0-5 and 
 0-3 per cent. It is clearly impracticable to have steel 
 standards with this gradation of carbon ; while, on the 
 other hand, to make a single colour comparable with 
 some one steel, and then by dilution make up others, 
 corresponding to all the desired percentages, is liable to 
 introduce an error. I have two steel bars that are kept 
 as standards, the carbon composition of which, as shown 
 by repeated combustions, is 0*501 per cent, and 0-304 
 per cent, respectively. As the limit of error in the 
 colorimetric test can be considered as about 0-01 per cent, 
 for this grade of steel, I am safe in calling the first a 0-5 
 and the second a 0-3 per cent, metal. From the 0-5 and 
 0-3 standard steels are now made up nitric acid solutions 
 under the same conditions, as regards amount of steel 
 taken, acid used, time of heating and temperature, as those 
 to which the regular tests are to be subjected. I am in 
 the habit of dissolving 0-5 grammes of drillings in 12 c.c. 
 nitric acid (1-20 specific gravity), and subjecting it in a 
 steam-bath to a temperature of 98 to 100 C. for fifteen 
 minutes. 
 
 'The next step is to make up from the chlorides a 
 colour which shall exactly match the 0'5 per cent, steel 
 solution, and another which, diluted to J- ths of its strength, 
 shall precisely correspond to the 0'3 per cent, steel. The 
 
384 ESTIMATION OF CAEBON IN STEEL. 
 
 former we will call a brown 0-5 colour, and the latter a 
 green 0-5 colour. I use the term " green 0'5 " for the 
 latter, because it will be found to be a colour which, 
 though of a greener cast, corresponds in intensity, before 
 diluting, to the solution of the 0-5 steel. These are pre- 
 pared as follows : 
 
 4 For the brown 0*5 colour, mix the chloride solutions 
 and water containing 0-8 per cent. (1-12 specific gravity) 
 hydrochloric acid in the ratio of 87 '5 c.c. water, 5*7 c.c. 
 cobalt chloride, 25 c.c. copper chloride, and 43 c.c. ferric 
 chloride. For the green 0'5 colour take 82-6 c.c. water 
 containing 0'8 per cent, hydrochloric (1-12 specific gravity), 
 5-4 c.c. cobalt, 5 c.c. copper, and 7 c.c. ferric chloride 
 solutions. The first of the colours so produced will ap- 
 proximately match the 0-5 steel solution, while the second, 
 diluted to J-ths of its strength with water containing 
 0*8 per cent, hydrochloric acid (1*12 specific gravity), will 
 approximately correspond to the 0-3 steel solution. The 
 absolute identity of the colours may be obtained by the 
 addition of the chloride or chlorides, the colour of which 
 corresponds to the shade that is wanting. The compari- 
 son of the green 0*5 colour with the 0-3 steel is readily 
 accomplished by the aid of a reading-tube suitably cali- 
 brated 6 c.c. and 10 c.c., for instance In the final match- 
 ing care should be exercised that the same light is used 
 as is to be regularly worked with the same window if 
 possible as this assures similarity of conditions as regards 
 transmitted and reflected light. With correct brown and 
 green 0*5 colours so obtained the alternate intermediate 
 points are easily acquired by mixing the two and diluting 
 with water containing 0'8 per cent, hydrochloric acid 
 (1-12 specific gravity) by the aid of a burette, as per 
 following table : 
 
DAY STANDARD COLOURS. 
 
 385 
 
 Per cent. Carbon 
 
 No. c.c. Brown 
 0'5 colour 
 
 No. c.c. Green 
 0'5 colour 
 
 No. c.c. Water 
 
 0-5 10 
 
 
 
 
 
 0-48 
 
 8-64 
 
 0-96 
 
 0-4 
 
 0-46 
 
 7-36 
 
 1-84 
 
 0-8 
 
 0-44 
 
 6-16 
 
 2-64 
 
 1-2 
 
 0-42 
 
 5-04 
 
 3-36 
 
 1-6 
 
 0-4 
 
 4 
 
 4 
 
 2 
 
 0-38 
 
 3-04 
 
 4-56 
 
 2-4 
 
 0-36 
 
 2-16 
 
 5-04 
 
 2-8 
 
 0-34 
 
 1-36 
 
 5-44 
 
 3-2 
 
 0-32 
 
 0-64 
 
 5-76 
 
 3-6 
 
 0-3 
 
 
 
 6 
 
 4 
 
 This table is based on the following considerations : 
 
 1. That theoretically, from a given solution represent- 
 ing carbon ol known value, other solutions representing 
 definite lower carbon percentages may be made by dilut- 
 ing with water. 
 
 2. That practically this will give colours of shades at 
 variance with those aimed at. 
 
 The explanation is clear. We have first to decide upon 
 some definite bulk as a suitable amount of the solutions 
 to make up. With the 5 inch by -| inch reading-tubes 
 adopted here, 10 c.c. has been found convenient. To 
 make the O48 standard, for instance, we must have such 
 a 0*5 colour that the shade produced upon diluting with 
 the proper amount of water will match that made from a 
 0-48 steel. This neither the green nor the brown 0-5 
 alone will do. The former is too green, the latter too 
 brown. The operation of determining in the mixture the 
 correct proportion of the two requisite to give the desired 
 colour (for the 0'48 colour as an example) is purely a 
 mathematical one. The computation is as follows : 
 Amount of green : amount of brown : : 0'02 : 0'18 ; 0-02 
 and 0-18 being the relative distances of the 0*48 colour 
 from the extremes, 0-5 and 0'3. The proper ratio of the 
 brown and green 0*5 colours being found, the amount of 
 the mixed chlorides and water to make a total of 10 c.c. 
 is easily calculated thus : 
 
 0-5 : 0-48 : : 10 : x 
 
 10-9-6 = 0-4 
 
 CC 
 
386 ESTIMATION OF CARBON IN STEEL. 
 
 Hence 9'6 will be the number of cubic centimetres of 
 the combined chloride solutions and 0-4 the number of 
 water. Any of the other gradations in this or any range 
 is found in a like manner. If it is desirable to make 
 either a higher or lower, all that is necessary is to accur- 
 ately know the carbon-contents of two suitable steels and 
 proceed as above. It is always well to keep the colours 
 from, exposure to the light as much as possible. When it 
 is necessary to use the colours constantly, as in testing 
 every heat at a Bessemer plant, a reading-rack capable of 
 being closed is to be recommended. This avoids the neces- 
 sity of removing to a dark place. Such a rack, provided 
 with a front and back door, swung on hinges fastened to 
 the bottom, and arranged with a thin plate of fine ground 
 glass for a reading background, has proved very effective. 
 For this purpose I have found the glass to possess advan- 
 tages over the mediums ordinarily employed. The solu- 
 tions must of course be corked tight. With due regard 
 to the precautions and directions here laid down, a set of 
 reliable inorganic standard solutions can be made from 
 the above chlorides that will be found to be constant for 
 four or five months of continued use, and often for even 
 a longer period. When decomposition does set in, the 
 colours gradually grow dark. 
 
 NIGHT STANDARD COLOURS. 
 
 As the method is commonly conducted, standards are 
 weighed from the steel direct. When tests have to be 
 made after every blow, this is obviously inconvenient, and 
 renders it unadvisable to leave the work in unskilled hands. 
 With the purpose of placing the night-tests as nearly as 
 possible on an equal footing with the day-tests, the follow- 
 ing investigation was undertaken by the writer. The day 
 chloride standards, when compared by both kerosene and 
 gaslight, were found to appear much lighter than the 
 corresponding steel solutions. The arc light gave the 
 same result. Monochromatic light as obtained by the 
 soda bead and various coloured glasses was tried without 
 
NIGHT STAND AED COLOURS. 387 
 
 success. The magnesium light gave the same reactions as 
 daylight, but was laid aside as impracticable, and the idea 
 of using the day-standards for night work was then aban- 
 doned. Night-standards that proved entirely satisfactory 
 were finally made up from the same solutions and on ex- 
 actly the same principle as the day-standards, with the 
 simple modification that the brown 0*5 and green 0'5 were 
 matched with the (>5 and O'o steel solutions in a dark 
 room, by the aid of kerosene light. To make the brown 
 0-5 night-standard, mix the solutions in the following 
 ratio: 7*4 c.c. cobalt, 1-5 c.c. copper, and 4*3 c.c, ferric 
 chloride solution with 86*8 c.c. water containing 0*8 per 
 cent, hydrochloric acid (1-12 specific gravity). For the 
 green 0*5 night-standard, mix in the ratio of 7*7 c.c. cobalt, 
 2*2 c.c. copper, and 4'5 c.c. ferric chloride solution with 
 85-6 c.c. water containing O8 per cent, hydrochloric acid 
 (1-12 specific gravity). The extremes being made up, the 
 intermediate colours are produced by the same table as 
 was used for the day -standards. The colour and intensity 
 of the brown and green 0*5 thus prepared will be approxi- 
 mately correct. The final shading must be accomplished 
 in the same manner as for the day-solutions. Colours so 
 produced will be found to be considerably darker than 
 the corresponding day- colours, when compared by day- 
 light. 
 
 For this work and for the regular night-tests I use a 
 rectangular camera, made of sheet iron, arranged on the 
 back with a door, open in front, and so cut, about midway 
 between -the ends, as to admit a light wooden rack holding 
 the colours. Three medium -sized kerosene-lamps, placed 
 on a stand in front, furnish the light. The rack is pro- 
 vided with a plate of ground glass as a reading back- 
 ground, and is so adjusted that a slide between the glass 
 and the lamps protects the standards from the light when 
 not in use, while the door on the back of the camera 
 serves to darken the other end. With standards made as 
 above, I have found no difficulty in turning out as accur- 
 ate work by night as by day, -and with the requisition of 
 no greater skill and ability than is required of any day 
 
 c c 2 
 
388 ESTIMATION OF SULPHUR IN IRON AND STEEL. 
 
 ca,rbon-boy. In these night-standards it will be noticed 
 that there does not exist that sharp contrast in shade 
 between the 0'3 colours produced by diluting the brown 
 and the green 0*5 as is found in the day-standards. It is 
 almost entirely a case of comparative intensity, when, un- 
 like the day-standards, the shade is of but little moment. 
 The kerosene light seems to eliminate to a great degree 
 the power of distinguishing variations in the green tint ; 
 and for this reason a very good set of night-colours could 
 be made by simply diluting the brown extreme with water 
 alone. It will not be as accurate a match, however, as if 
 the green is used also. It takes a little time to familiarise 
 the eye to night-reading ; but when once accustomed it 
 has been my experience that as close, if not closer, discri- 
 minations can be made between the colours by night as by 
 day. In comparing by dilution, the failure to take into 
 consideration the distinction of shade as well as intensity 
 has often been the source of considerable error. With 
 standard solutions made as directed, this liability is reduced 
 to a minimum ; and this, with the manifest advantages of 
 ease and celerity, renders inorganic standards a desidera- 
 tum for any laboratory dealing with continuous work of 
 this kind. 
 
 Estimation of Sulphur in Iron and Steel. The plan 
 usually adopted is to dissolve a weighed quantity (about 
 3 grms.) of the metal in strong nitric acid, adding a little 
 hydrochloric acid occasionally, and evaporating the solu- 
 tion to dryness. Dissolve the residue in very dilute warm 
 hydrochloric acid, and precipitate the sulphuric acid in 
 the solution by means of chloride of barium. If the pre- 
 cipitated barium sulphate, after washing once or twice by 
 decantation, has a yellow or brown colour, owing to the 
 presence of iron mechanically carried down, heat it, before 
 filtering, with dilute hydrochloric acid. 
 
 Dr. Eggertz, to whom analytical chemistry is indebted 
 for the colorimetric process of estimating combined carbon 
 in iron and steel, has devised a very expeditious plan 
 for estimating the sulphur. He takes one decigramme of 
 cast iron, wrought iron, or steel, cut up or pulverised, and 
 
EGGERTZS METHOD. 889 
 
 passed through a sieve with holes not larger than 0*6 m.m., 
 and introduces it, by means of a glass or glazed-paper 
 funnel, into a flask about 0-15 metre high and 5 centimetres 
 diameter, previously containing 1 grm. of water and 0*5 
 grm. of concentrated sulphuric acid ; or, in preference, 
 1-5 grin, of sulphuric acid, sp. gr. 1-25, and whose volume 
 (1-5 c.c.) has been marked on the flask. A piece of 
 polished silver plate (18 m.m. long, 7*5 m.m. wide, and 1 
 m.m. thick, with a hole at one end), composed of 75 per cent, 
 of silver, 25 per cent, of copper, and attached to a thin 
 platinum or silver wire, is quickly introduced into the 
 flask, so that it may be a little below the neck ; a cork is 
 put in so as to hold the wire without completely closing it. 
 It is allowed to stand fifteen minutes at the ordinary tem- 
 perature, and the silver plate is then removed. If the iron 
 contains sulphur, the plate is coloured by the sulphuretted 
 hydrogen gas disengaged during the solution of the iron 
 in the dilute sulphuric acid ; and, according to the amount 
 of sulphur present, the coloration of the plate passes to 
 a coppery yellow, a bronze brown, a bluish brown, or a 
 blue. These colorations are estimated with the greatest 
 accuracy especially that of the silver plate alone, No. 1 ; 
 that of coppery yellow, No. 2 ; that of bronze brown, No. 
 3 ; that of blue, No. 4. The intermediate degrees may be 
 represented by decimals, thus : 2 ! 5 if the coloration is 
 between 2 and 3; 3 p l if the plate is one-tenth towards the 
 blue ; 3*5 if it is as much blue as brown ; 3*9 if the brown 
 coloration is feeble. 
 
 As the normal coloration, No. 2, Dr. Eggertz has 
 adopted that of the bronze called yellow metal, newly rubbed 
 with fine sand on leather. (This metal consists of 60 parts 
 of copper and 40 of tin.) For the coloration No. 3, a 
 convenient alloy has not yet been found. A bronze, con- 
 sisting of 85 parts of copper and 15 parts of zinc, does not 
 quite represent the colour which should be obtained, for 
 when freshly cleaned it is too bright, and at last takes a 
 bluish coloration'. For the colour in question it is better 
 to use a plate of silver which remains in the flask during 
 the solution of the iron, until it has become as brown as 
 
390 ESTIMATION OF. SULPHUR IN IRON AND STEEL. 
 
 possible,- and a slight bluish colour begins to be perceived ;, 
 the plate is then removed and preserved in a well-closed 
 tube. The colour No. 4 resembles that of a watch-spring. 
 If the amount of sulphur is very considerable, this colora- 
 tion passes to a clear bluish grey. By passing the plate of 
 silver over a bottle of sulphide of ammonium the desired 
 number can be easily obtained. To obtain in these assays 
 for sulphur the proper tint on a silver plate, it is necessary 
 to take certain precautions. The plate is to be held in 
 pincers and cleaned as well as possible by rubbing it with a 
 soft leather on which is placed a little fine rotten-stone. 
 Contact with the fingers is avoided by means of a piece of 
 paper, and the plate is to be carefully dried with a piece of 
 filtering-paper. If the plate, by cleaning or by the action 
 of the burnisher, has been purified on its surface, this should 
 be carefully removed by again rubbing with the leather, for 
 pure silver is less sensitive to the action of the gas than that 
 of the given standard. Thus it has sometimes been found 
 that the silver employed for coinage furnishes less homoge- 
 neous plates, of which those parts richest in copper assume 
 more quickly the blue coloration. On this account, the 
 plates should be compared between themselves, by intro- 
 ducing them at the end of a wire into a flask in which iron 
 is dissolving containing from 0-05 to 0-08 per cent, of sul- 
 phur. On introducing the plate, care must be taken not to 
 turn the side but the edge against the strongest current of 
 gas, whicn would otherwise colour one face of the plate 
 stronger than the other. The plate should be rapidly in- 
 troduced into the flask after the introduction of the iron, as 
 then a very strong disengagement of sulphuretted hydrogen, 
 immediately takes place. After a first experiment the flask 
 is to be filled with water several times, so as to get rid of 
 the odour of sulphuretted hydrogen. If a steel mortar is 
 employed to pulverise the iron, the whole of the piece 
 selected should be reduced to a very fine powder. The 
 mortar should be well cleaned each time, taking care to 
 remove the disc from the bottom. Changes in temperature 
 between 15 and 25C. seem to have no sensible influence on 
 the coloration of the metal ; if the temperature exceeds. 
 
EGGERTZS METHOD. 391 
 
 40 the plate becomes moist and gives false indications. 
 Some practice is required to judge of the coloration of 
 the plate, but it may be easily acquired. Generally the 
 best plan is to place the standard plates of tints 1, 2, 3, 
 &c., on a sheet of white paper by the side of the plate 
 under experiment, exposing them to a good light near a 
 window (but not sunlight), and to examine them with a 
 lens. The colorations between 2 and 4 are the most 
 difficult to recognise ; but with a little experience none 
 will vary more than Ol ; so that, for instance, the colora- 
 tion may be estimated between 3'5 and 3 '6. 
 
 The following is somewhat an approximation between 
 the different colorations upon the silver plate and the 
 amount of sulphur in a great number of different samples 
 of iron : 
 
 Number of Percentage 
 
 for Coloration of Sulphur 
 
 1-0 0-00 
 
 1-2 0-01 
 
 2-0 0-02 
 
 2-5 0-03 
 
 3-0 0-04 
 
 3-1 0-05 
 
 3-2 0-06 
 
 3-3 0-07 
 
 3-5 0-08 
 
 3-6 0-09 
 
 3-7 0-10 
 
 3-8 0-12 
 
 3-9 0-15 
 
 4-0 0-20 
 
 It is evident that in this way the exact quantity of the 
 sulphur is not estimated ; but several years' experience 
 has shown that if these experiments are made with care, 
 and the quantity of sulphur does not exceed O'l percent., 
 the results are near enough for all practical purposes. 
 Iron which does not colour the silver plate will sometimes 
 produce a coloration if we double the quantities of iron 
 and acid. With half the quantities of acid and sulphurised 
 iron, silver generally gives a little more than half the real 
 quantity of the sulphur which is present. 
 
 Amongst experiments on the estimation of sulphur in 
 iron, the following deserve mention : 1. The quantity of 
 sulphur in wrought iron is often so small that it produces 
 
S92 ESTIMATION OF SULPHUR IN IRON AND STEEL. 
 
 no coloration on the silver plate ; this iron, therefore, not 
 being red-short, may be employed for all kinds of uses. It 
 must not, however, be forgotten that the quantity of sul- 
 phur is not equally distributed throughout a piece of iron, 
 but that it may vary considerably in different places. On 
 experimenting with the turnings obtained from a portion 
 of an iron bar which was visibly red-short, a stronger tint 
 is often obtained on the silver plate than when using other 
 parts of the bar. The fragments obtained from red-short 
 iron in boring a horse-shoe do not often give on the silver 
 plate a deeper coloration than 2, and it appears to follow 
 that ordinary wrought iron which contains 0'02 per cent, 
 of sulphur in certain parts cannot conveniently be em- 
 ployed for this purpose. If the red-short iron gives to the 
 plate a slighter and more feeble coloration than 2, it may 
 be supposed that the breaking is due less to sulphur than 
 to an insufficient working of the cast iron, the crude pieces 
 in wrought iron entirely free from sulphur often acting as 
 if they were red-short. In general it appears certain that 
 the quantity of sulphur in iron is more injurious when the 
 iron has been badly worked. In a hard iron melted in a 
 steel crucible we may, in spite of its containing O04 per 
 cent, of sulphur, make holes like those in a horse-shoe 
 without any trace of cracks, which may undoubtedly be 
 attributed to the homogeneity and good working of this 
 iron ; the quantity of phosphorus being only 0'3 per 
 cent. The lower portion of an English rolled rail, without 
 fault, contained 0*11 per cent, of sulphur and O03 per cent, 
 of phosphorus ; a portion was cut off which was so red- 
 short that it could not be made use of. 
 
 2. The amount, in sulphur, of steel of the best quality 
 is such that the colorations on the silver plate vary only 
 between 1 and 1'5. As in the case of wrought iron, the 
 quantity of sulphur often varies in different parts of the 
 same piece of steel, and that also appears to be the casein 
 a little less decided manner in cast steel. 
 
 3. The quantity of sulphur in cast iron is rarely so little 
 as not to colour the plate. In the greater number of 
 Swedish cast irons, this quantity is such that the silver 
 plate varies in coloration between 2 and 3. In iron for 
 
NEW COLORIMETE1CAL PROCESS. 30:} 
 
 gun-castings it is between 3'3 and .3*7, and sometimes 
 more. In cast iron the quantity of sulphur is often dis- 
 tributed unequally ; there is generally more on the surface 
 than is met with below. If the coloration of the silver- 
 plate does not- exceed 3, we can assume that the cast iron 
 refined in the ordinary manner will not give red-short 
 iron, especially if the refining is done carefully. But as in 
 different methods of refining, different quantities of sulphur 
 may be removed from the iron, and in general more if the 
 iron is the result of a light charge of the blast furnace, it 
 cannot be said beforehand that all cast iron which commu- 
 nicates a bluish coloration to the plate will necessarily give 
 red-short iron. This will be the case, however, with cast 
 iron which colours the plate as deep a blue as that of 
 a watch-spring. In cast iron, which gives a red-short 
 wrought iron, without rendering the silver plate more 
 than brown, it is probable (the iron having been well re- 
 fined) that the cause is owing to the presence of other 
 substances than sulphur ; but this occurrence is very rare. 
 Many circumstances appear to show that the quantity 
 of sulphur in iron diminishes with time, at least on the 
 surface, and under favourable conditions. 
 
 New Colorimetrical Estimation of Sulphur in Iron. 
 
 The above-described colorimetrical method of esti- 
 mating the amount of sulphur in iron, worked out by Prof, 
 v. Eggertz, has been, and still is, indisputably of great 
 utility to the Swedish iron manufacture. The method 
 since then has come into general use in our ironworks. 
 Mr. J. Wiborgh sought to perfect such a colorimetrical 
 process as would be applicable also for greater amounts 
 of sulphur than that for which the Eggertz method is in- 
 tended. This new colorimetrical sulphur test fulfils the 
 demands that in general may be put upon the practical 
 requirements intended in the method namely, it is quick 
 and easy to execute ; also, it has sufficient accuracy to give 
 the amount of sulphur, whether in cast iron, steel, or 
 wrought iron. The following is Mr. Wiborgh's description 
 of the method. 
 
394 ESTIMATION OF SULPHUR IN IKON AND STEEL. 
 
 Basis of the Method. 
 
 Iron dissolves completely in diluted sulphuric acid 
 or hydrochloric acid, and the gases evolved hydrogen, 
 carburetted hydrogen, and sulphuretted hydrogen pass 
 through a cloth impregnated with a metallic salt, and 
 through the action of sulphuretted hydrogen a metallic 
 sulphide is formed, which colours the cloth. From the 
 intensity of the colour afterwards the amount of sulphur 
 in the iron is decided. 
 
 I proceed here upon the assumption that a given sur- 
 face is always coloured equally strongly by a fixed quantity 
 of sulphur. But to obtain an equal quantity of sulphur 
 from two specimens of iron which have unequal amounts 
 of sulphur, evidently the amounts of iron weighed out 
 must be inversely proportionate to the amounts of sulphur. 
 Therefore a w r eight W of an iron with sulphur amount S 
 gives the same colour as a weight W of another iron with 
 sulphur amount S' ; so must 
 
 WS = WS'; 
 and as S 7 is the amount of sulphur sought, 
 
 WS 
 
 S'-'w 7 '- 
 
 If you have thus an iron (normal iron) with the amount 
 of sulphur accurately known, you can, by varying the 
 amounts of it weighed out, produce a colour-series in 
 which for every colour you know the product W S, 
 This colour-series constitutes a scale with the aid of which 
 the unknown amount of sulphur S' in another iron can be 
 estimated by dividing the same colour's known product, 
 W S, by the weight of the iron used for the assay. 
 
 The apparatus shown in the accompanying figure con- 
 sists of a small boiling flask, A, provided with a close-fitting 
 caoutchouc stopper, m, in which are placed a glass cylin- 
 der, R, with one end drawn out to a tube, p, and the other 
 end having a flat polished flange, G, also a funnel tube, T, 
 to introduce the acid. This latter consists properly of 
 two, T and T', united with a caoutchouc tube, over which 
 
NEW COLOR IMETRICAL PROCESS. 
 
 395 
 
 FIG. 103. 
 
 is placed a nipper tap, K, with a screw, so that the quan- 
 tity of acid to be passed into the flask may be accurately 
 regulated. Upon the cylinder flange is placed a caoutchouc 
 ring, N, and upon this the prepared cloth o. That the 
 ring may close tightly against the flange and cloth, lay 
 upon the cloth another caoutchouc ring, N', of the same 
 size as the former, and outermost a wooden ring, s, which 
 is pressed against the flange by the spring B. 
 
 These caoutchouc rings ought to have a precise inside 
 diameter, for upon the size depends how much of the sur- 
 face of the cloth will be coloured ; 
 as showing, by the drawing, it is 
 made less than the opening of 
 the cylinder, because it is easier 
 to obtain such rings of a precise 
 size than the glass cylinder. In 
 the apparatus represented the 
 diameter of the ring o is 55, and 
 the opening of the cylinder 58 
 millimetres. 
 
 In order that no steam may 
 condense upon the ring s and 
 run down to the cloth, it is best 
 to make this ring of wood rather 
 than of glass or metal, of the 
 size and form shown in the 
 drawing. 
 
 By the arrangement now 
 described neither gas nor steam 
 can come from the apparatus 
 without first passing through the cloth. 
 
 To heat the flask to boiling it is placed upon a sand- 
 bath, E, which rests upon the stand D, and is heated by 
 means of a gas or spirit lamp, p. 
 
 Preparation of the Cloth. 
 
 In the beginning I used, instead of cloth, unglazed 
 paper, such as filter-paper ; but I soon found that cloth 
 
396 ESTIMATION OF SULPHUR IN 'IRON AND STEEL. 
 
 was much more serviceable. Such paper certainly per- 
 mits steam and gas to pass ; but it is tender, and easily 
 breaks asunder with the least incaution while boiling. 
 
 The cloth used is common, fine, white cotton calico. 
 Linen is less suitable, for it is thin, and therefore does not 
 absorb so easily all the sulphuretted hydrogen as evolved. 
 The preparation of the cloth is simply to moisten it with 
 a solution of a metal salt. For this may be employed 
 either lead, silver, copper, cadmium, or antimony salts ; 
 but of all these I have found cadmium salts the most 
 serviceable. 
 
 Silver and lead salts are certainly very sensitive to 
 sulphuretted hydrogen ; but the combinations of these 
 metals with sulphur are black, and colour the cloth too 
 strongly, so that it will be necessary to employ small 
 weighings of the sample or extravagantly large apparatus. 
 
 Copper salts certainly give brownish and considerably 
 softer colours ; but, in consequence of the property of the 
 sulphide to enter into variable combinations with the oxide, 
 they are easily oxidised, and not permanent. 
 
 Antimony has few soluble salts ; and, besides, in one case 
 these are seen to be less sensitive to sulphuretted hydro- 
 gen : they are little fit for this purpose. A sufficiently 
 dark colour is obtained by mixing the salts as, for ex- 
 ample, cadmium and lead salts but has not led to any 
 good results, as the colour comes out uneven, darker upon 
 some places than upon others, so that the cloth has a 
 more or less flannel-like aspect, varying between yellow 
 and black. 
 
 Eor these reasons I have selected cadmium salt alone 
 for the preparation of the cloth, and this the rather as the 
 sulphur combination of cadmium is a particularly beau- 
 tiful and constant yellow colour ; also, that the affinity of 
 this metal for sulphur is so great that it surpasses that of 
 lead. 
 
 When the cloth is impregnated with a mixture of cad- 
 mium and lead acetates, equal parts of each, and then 
 exposed to a small quantity of sulphuretted hydrogen, 
 the cloth at first will be almost exclusively yellow, and 
 
NEW COLORIMETRICAL PROCESS. 397 
 
 by the passage of more sulphuretted hydrogen it becomes 
 dark, which shows well that the gas is sooner decom- 
 posed by cadmium than by lead. Again, concerning which 
 of the cadmium salts ought to be selected for impregna- 
 tion of the cloth it appears to be a matter of indifference 
 which is employed, although the colour tint will be some- 
 what different as one or other salt is used. 
 
 Cadmium nitrate gives a singularly high and strikingly 
 beautiful orange colour ; the sulphate is somewhat weaker, 
 with brownish yellow tint, while the chloride and acetate 
 give lighter colours. 
 
 The colours occasioned by the varying amounts of sul- 
 phur may perhaps be most easily judged separately when 
 the cloth is prepared with cadmium nitrate ; but not- 
 withstanding I have likewise used the acetate, because 
 I think it likely that this salt, which has the weakest 
 and most volatile acid, ought to give colours that would 
 be least changeable, which property is of considerable 
 weight. 
 
 Tn order that all the sulphuretted hydrogen may be 
 absorbed by the cloth, it must contain a certain amount 
 of cadmium salt in proportion to the largest quantity of 
 sulphur which may possibly be present ; otherwise a por- 
 tion of the sulphuretted hydrogen will pass through the 
 cloth, colouring yellow not only the under side but also 
 the upper side. 
 
 A solution of 5 grms. crystallised cadmium nitrate in 
 100 c.c. distilled water is of suitable strength. 
 
 Cloth of the proper fineness, prepared with such 
 solution, allows not a trace of sulphuretted hydrogen to 
 pass through, for the cloth was coloured only on the 
 under side, and when double folds were used the upper 
 had not the faintest colour. 
 
 The general influence of the degree of concentration of 
 the solution is that the colours from strong solutions lie 
 more upon the surface of the cloth, and that of weaker 
 solutions penetrates deeper into the cloth ; and this 
 causes a certain unevenness in the aspect of the colour, 
 although the amount of sulphide of cadmium in both 
 
398 ESTIMATION OF SULPHUR IN IRON AND STEEL. 
 
 cases is alike. The influence of a smaller variation in 
 the degree of concentration is not perceptible. 
 
 The cloth is prepared by cutting, on a round model, 
 several folds to about 80 m.m. diameter. These are laid 
 in the solution of cadmium acetate, care being taken that 
 each circle is thoroughly saturated by the solution. After 
 some minutes take out the pieces, and spread them out 
 on a clean cloth until completely dried ; then place them 
 for safety in a suitable box. 
 
 The Colour Scale. 
 
 According to the quantity of sulphuretted hydrogen 
 which reacts upon the prepared cloth, so is it covered with 
 different amounts of cadmium sulphide, and receives a 
 more or less yellow colour. The sensibility in this respect 
 is so great that indeed T oVo tn P art f a ni.grm. of sulphur 
 is able to communicate to the surface of a square centi- 
 metre of the white cloth a certainly weak but yet very 
 manifest yellow colour. Increase successively the amount 
 of sulphur by T^OO^ P art f a m -g rm - per square centi- 
 metre ; you obtain by this small addition of sulphur from 
 the first a clear distinction in the intensity of the. colour ; 
 but in proportion as the strength of the colour increases 
 with the augmented amounts of sulphur, the difference 
 will be all the more difficult to observe. When the amount 
 of sulphur increases to about 0'02 m.grm. per square c.m., 
 the cloth is now strongly coloured, and to produce a 
 manifestly distinct difference of colour intensity it requires 
 two or three times greater excess of sulphur than with the 
 lower colour estimations. Consequently you cannot with 
 advantage make use of very strong colours, because the 
 difficulty of their comparison is considerably increased. 
 
 In order to estimate the amount of sulphur in an iron, 
 you must first have a colour series, or scale of colours, 
 whereon every colour number represents a certain amount 
 of sulphur per cent., presupposing that a certain quantity 
 of iron has been weighed out for assay. 
 
 This colour series can be easily procured by the help of 
 
NEW COLORIMETRICAL PROCESS. 399 
 
 a normal iron in which the amount of sulphur has been 
 estimated beforehand with the greatest accuracy. How 
 great the amount of sulphur this normal iron may con- 
 tain is quite immaterial, but use the most trustworthy 
 among the known methods of estimating sulphur in iron 
 by oxidation of the sulphur, and thereafter precipitating 
 by chloride of barium. With small amounts of sulphur 
 this is to a certain degree unsafe, so it is better to use for 
 the scale an iron with a pretty high amount of sulphur, 
 such as about O'l per cent., because the influence of 
 a small error in this estimation of the sulphur will be 
 less. 
 
 If you have a known amount of sulphur S, also an iron 
 of which 0'4 grm. is weighed out, and its amount of sul- 
 phur S', to obtain the same colour from both of the irons 
 
 must be 
 
 w x s = 0-4 x S' 
 
 Or 
 
 w _0'4xS / . 
 
 In this formula place instead of S' the successive 
 amounts of sulphur per cent., as 0*005, Q-01, 0-02, &c. As 
 you know how much of the normal iron in every case 
 ought to be weighed, so you have a series of colours that 
 are entirely the same as those which will be obtained if 
 you had different irons with their respective amounts of 
 sulphur and 0'4 grm. of each weighed ; you get, in other 
 words, a colour scale, where colours give directly the 
 amount of sulphur the iron contains. 
 
 Make up such a scale for the apparatus shown in the 
 sketch, with the inner ring 55 m.m. diameter, under the 
 supposition that 0'4 grm. has been weighed for testing. 
 It is not desirable to go farther than this, so that the 
 highest colour number corresponds to 0*1 per cent, sul- 
 phur, because otherwise the colours will be too strong. 
 Nevertheless, you must not take between the high colour 
 numbers a greater difference in the amount of sulphur 
 than between the lower, if the different colour intensities 
 are to be clearly separated from each other. 
 
400 ESTIMATION OF SULPHUR IN IRON AND STEEL. 
 
 For example, you can allow the scale to be composed 
 of seven colour numbers, taken thus 
 
 No. 1 corresponding to 0*005 per cent sulphur. 
 
 2 0-01 
 
 3 0-02 
 
 4 0-03 
 
 5 0-05 
 
 6 0-07 
 
 7 0-10 
 
 If thus 0-4 gramme iron is weighed for assay, one can, 
 with the help of the above scale, estimate the amount of 
 sulphur up to O'l per cent. The accuracy wherewith the 
 estimations are done is, for the lower of amounts of sul- 
 phur at least, equal to 0*005 per cent., and for higher O'Ol 
 per cent. ; for the differences between the colours of the 
 scale are here so great that one with the greatest ease can 
 estimate a colour lying between two colour numbers. 
 
 The same scale may, however, be used to estimate 
 quickly whatever amount of sulphur is desired, if only the 
 weight of iron used for assay be varied. It is obvious 
 that absolutely the same colour which an iron gives from 
 0'4 gramme must be got from another iron with half as 
 much sulphur if 0-8 gramme has been weighed ; and that, 
 in general, if 0'4 n is weighed, the colour scale comes to 
 represent sulphur amounts which are J of those that 
 answer to the weight 0'4 gramme. 
 
 One ought therefore to use in general greater amounts 
 for assay, when low amounts of sulphur are to be esti- 
 mated with great accuracy, also lesser quantities for high 
 amounts of sulphur. 
 
 In case the colour comes out too strongly lying upon 
 the margin or beyond the greatest amount of sulphur 
 in the scale, the assay ought to be made with a lesser 
 weight. 
 
 To avoid calculation one can, under every colour num- 
 ber, place not only the amount of sulphur which corre- 
 sponds to that weighed out in making the scale, but also 
 that which corresponds to some other weighings which 
 possibly may occur. Note the following weighings, W, 
 also the corresponding amounts of sulphur, S, per 
 cent. : 
 
NEW COLORIMETKICAL PROCESS. 401 
 
 0-8 gramme. 0-005 per cent. 
 
 0-4 , 0-01 
 
 0-2 
 
 0-1 
 
 0-08 
 
 0-04 
 
 0-02 
 
 0-02 
 
 0-04 
 
 0-05 
 
 0-1 
 
 0-2 
 
 With these different weighings are thus given the colour 
 numbers of the amounts of sulphur from 0-005 per cent, 
 to 0-2 per cent. 
 
 Having by the above method prepared the different 
 colour numbers which form the scale, they may be arranged 
 in order upon small white drawing-paper, bound, and pre- 
 served in a suitable portfolio. 
 
 When cadmium acetate is used for the preparation of 
 the cloth, the colours will be very constant. They have 
 undergone no sensible change for several months, though 
 preserved only in the above manner. One such scale 
 ought to be made for every apparatus, and observe that 
 the cloth used for the preparation of the scale must be the 
 same sort as that afterwards used for testing. 
 
 Details of the Process. 
 
 Rinse every part of the apparatus with water, so that 
 no acid remains from the former assay. The boiling-flask 
 is half filled with distilled water ; then fit the apparatus 
 together and place on the sand-bath, which is heated by 
 means of a gas- or spirit-lamp, so that the water in the 
 flask comes to gentle boiling. While heating, weigh out 
 the iron to be tested. This may certainly be in the form 
 of small pieces, but, that the solution may not proceed too 
 slowly, it is best for the iron to be in the form of filings, 
 turnings, or powder. Difficultly soluble iron, such as white 
 iron high in silicon, phosphorus chrome iron, &c., ought 
 always to be finely pulverised. 
 
 The weighed sample is brought, by means of a small 
 funnel and hair-pencil, into the test-tube r. This test- 
 tube has the mouth widened. It is set in a loop made on 
 the end of a platinum wire w, which is bent over the 
 edge of the test-tube as shown in the figure, so that the 
 platinum wire may not slip over the tube and rest on the 
 
 D D 
 
402 ESTIMATION OF SULPHUR IN IKON AND STEEL. 
 
 bottom of the flask ; this ought to be avoided, because 
 the test-tube may take such a position that the access of 
 the acid to the iron maybe made difficult. The platinum 
 with which the test-tube is thus fastened ought to be 
 rather stiff about 0'3 m.m. diameter. 
 
 After the water has boiled two minutes, and the air 
 has been expelled, take out the stopper with the cylinder 
 attached. The test-tube with the sample is lowered into 
 the flask, where it rests on the bottom, being held upright 
 by the platinum wire. 
 
 Put the apparatus together again, and upon the 
 cylinder flange lay the above-mentioned caoutchouc ring, 
 55 m.m. in diameter, and over it the prepared cloth, 
 another caoutchouc ring, and lastly the wooden ring, all 
 of which are pressed together by the spring clamp. 
 
 As soon as the cloth is laid on screw up the clamp K on 
 the funnel-tube, and steam must now pass through the cloth. 
 
 Partly to get the air as much as possible driven out of 
 the apparatus, and partly to moisten the cloth, maintain 
 the water in gentle boiling for about eight to ten minutes 
 before any acid is introduced. Then fill the funnel-tube 
 with diluted sulphuric acid (for example, J volume sul- 
 phuric acid, 1*83 sp. gr., and |- volume water). Open care- 
 fully the screw-clamp K, and allow the acid to drop into the 
 flask. For 0-4 gramme iron use about 10 c.c. dilute acid. 
 
 As soon as acid comes down, the solution of the iron 
 begins ; steam and gases pass through the cloth, and in 
 proportion as the solution proceeds, also according to the 
 amount of sulphur in the iron, the under side of the cloth 
 is more strongly coloured yellow. 
 
 After all the iron is dissolved, boil the solution further 
 from five to ten minutes, to drive out the sulphuretted 
 hydrogen which yet remains in the apparatus. Then open 
 the spring clamp, remove the bottom ring, and lay the 
 cloth upon a piece of filter-paper to dry completely. It 
 only remains now to estimate the amount of sulphur by 
 comparing the colour of the cloth with the colour scale. 
 
 That, the liquid in the flask during the whole period be 
 maintained at an even but gentle boiling is very essential 
 
NEW COLORIMETRICAL PROCESS. 403 
 
 for this process. The boiling ought to be so strong that 
 the steam is always seen to pass through the cloth, 
 but by no means so violent that the cloth becomes 
 stretched by the boiling. For when it happens that the 
 cloth, in consequence partly of this stretching and partly 
 from the condensed steam, becomes more and more close, 
 and takes a strongly convex form, then the pressure in the 
 apparatus increases, so that on the introduction of the acid 
 the gases formed rush through the funnel-tube. With 
 cautious boiling one need never fear such an occurrence. 
 
 In order to avoid oxidation of the sulphuretted hydro- 
 gen formed during the solution of the iron, it is of 
 importance that before the acid is introduced into the 
 flask the air be as completely as possible expelled from the 
 water and the apparatus, also that the boiling be so strong 
 that plenty of steam always accompanies the gases which 
 pass through the cloth. 
 
 The cloth in the assay ought to be coloured evenly, for 
 if the colour is uneven it very considerably increases the 
 difficulty of estimating the colour strength. 
 
 Whether the colour comes out even or not depends 
 chiefly upon the construction of the glass cylinder. It 
 ought to be so accurately made that the centre line of the 
 tube coincides with that of the cylinder. Further, the 
 tube ought to be short and of a conical form, 7 to 8 m.m. 
 diameter at the lower end. 
 
 If the tube be too wide the cloth is always coloured un- 
 evenly ; and again, if it be too small, drops of water con- 
 densed in the tube are cast up on the cloth, which give 
 it a spotted aspect. The drawing represents the cylinder 
 so set in the caoutchouc stopper that the mouth of the 
 tube is even with the under side of the stopper ; but it is 
 still better to allow the end of the tube to be 5 to 10 m.m. 
 under the stopper. It is certainly now more difficult for 
 the water to leave the tube, and causes especially at 
 first, before the cylinder becomes warm a weak bubbling ; 
 but this has no hurtful influence, but contributes to the 
 more even coloration of the cloth. 
 
 Observe, further, that the glass cylinder be placed by 
 
 D D 2 
 
404 
 
 ESTIMATION OF SULPHUR IN IRON AND STEEL. 
 
 the eye as vertical as possible, also that the apparatus be 
 riot placed in a draught. 
 
 The time required for a sample is from half to three- 
 quarters of an hour, according as the iron is more or less 
 soluble. 
 
 This new method, in many special experiments, has given 
 most satisfactory results, in that it is independent of the 
 amounts of carbon and silicon in the iron, indicating high 
 as well as low amounts of sulphur in iron, as seen in the 
 accompanying table, with sulphur estimations in different 
 sorts of iron. These have been performed by this method, 
 and controlled by accurate estimations of the sulphur by 
 the wet method. 
 
 No. 
 
 Sorts of Iron tested for Sulphur by 
 Wiborgh's Colorimetric Method and by 
 Chloride of Barium 
 
 By 
 
 Wiborgh's 
 Method. 
 Sulphur, 
 p.c. 
 
 By Wet Method with Chloride 
 of Barium 
 
 .A. 
 
 Sulphur, 
 p.c. 
 
 Estimated by 
 
 1 
 
 White charcoal cast iron 
 
 0-005 
 
 0-005 
 
 A. Tamm. 
 
 2 
 
 Spiegeleisenfrom Siegen,with 
 
 
 
 
 
 0*05 per cent, copper . 
 
 0-005 
 
 0-006 
 
 J. W T iborgh. 
 
 3 
 
 Bar iron, with 0*076 per cent. 
 
 
 
 
 
 arsenic . . . . 
 
 0-007 
 
 0-008 
 
 Y. Lagervall. 
 
 4 
 
 Casting 
 
 0-0075 
 
 0-005 
 
 A. Tamm. 
 
 5 
 
 Grey charcoal cast iron 
 
 0-0075 
 
 0-005 
 
 A. Tamm. 
 
 6 
 
 White 
 
 0-012 
 
 0-01 
 
 J. Wiborgh. 
 
 7 
 
 Casting ..... 
 
 0-012 
 
 0-014 
 
 A. Tamm. 
 
 8 
 
 Half- white charcoal cast iron 
 
 0-018 
 
 0-018 
 
 J. Wiborgh. 
 
 9 
 
 Malleable casting . 
 
 0-015 
 
 0-013 
 
 N. Lagerfelt. 
 
 10 
 
 White charcoal cast iron,with 
 
 
 
 
 
 0*071 per cent, arsenic 
 
 0-02 
 
 0-025 
 
 Y. Lagervall. 
 
 11 
 
 White charcoal cast iron 
 
 0-02 
 
 0-02 
 
 J. Wiborgh. 
 
 12 
 
 Malleable casting . 
 
 0-023 
 
 0-024 
 
 E. V. Zweigbergk. 
 
 13 
 
 White charcoal cast iron 
 
 0-023 
 
 0-024 
 
 Y. Lagervall. 
 
 14 
 
 ?> > 
 
 0-025 
 
 0-022 
 
 A. Tamm. 
 
 15 
 
 > J> JJ 
 
 0-028 
 
 0-029 
 
 J. Wiborgh. 
 
 16 
 
 Cast iron melted with copper, 
 
 
 
 
 
 containing 1'55 p.c. Cu 
 
 0-039 
 
 0-038 
 
 5 
 
 17 
 
 Siemens-Martin iron 
 
 0-04 
 
 0-037 
 
 >5 
 
 18 
 
 5> >5 
 
 0-05 
 
 0-047 
 
 }> 
 
 19 
 
 White charcoal cast iron,with 
 
 
 
 
 
 0-015 per cent. Cu 
 
 0-06 
 
 0-061 
 
 JJ 
 
 20 
 
 Steel 
 
 0-07 
 
 0-068 
 
 
 21 
 
 Siemens-Martin iron . 
 
 0-1 
 
 0-093 
 
 1J 
 
 22 
 
 Grey cast iron 
 
 0-135 
 
 0-134 
 
 Y. Lagervall. 
 
 23 
 
 Cannon cast iron . 
 
 0-15 
 
 0-145 
 
 J. Wiborgh. 
 
 24 
 
 White cast iron from Horde, 
 
 
 
 
 
 with 1-88 per cent. P . 
 
 0-21 
 
 0-19 
 
 E. Aquilon. 
 
 25 
 
 Malleable casting . 
 
 0-35 
 
 0-34 
 
 J. Jungner. 
 
 26 
 
 WTiite cast iron from coke . 
 
 0-45 
 
 0-46 
 
 J. Wiborgh. 
 
 27 
 
 " 
 
 0-7 
 
 0-66 
 
 P. G. Lidner. 
 
KEW COLORIMETRICAL PROCESS. 405 
 
 I have also tested if some impurities in iron such as 
 copper and arsenic had any influence upon this method. 
 For this purpose several sorts of iron containing copper 
 and arsenic were specially selected from the collection of 
 the Mining School ; and in the table it is found 
 
 I. In the samples 2, 16, and 19 copper has no action ; 
 also 
 
 II. In samples 3 and 10 the same is probably the case 
 with arsenic, for the certainly somewhat large difference 
 in the sulphur estimation which occurs in sample 10 
 must be sought in other circumstances, as sample 3 even 
 with its small amount of sulphur and high amount of 
 arsenic shows no noticeable difference. 
 
 The apparatus may be made of whatever size is 
 wished, but in relation to use in Sweden it is seldom 
 necessary to examine iron with a greater amount of sul- 
 phur than O'l per cent. I regard an apparatus with an 
 inner ring of 55 in.m. as diameter of a suitable size. On 
 the other hand, if one has in general to test iron with 
 high amounts of sulphur, as cast iron made with coke, 
 &c., I would recommend that the apparatus be made 
 somewhat larger. 
 
 Estimation of Sulphur in Iron Ores. Five grammes 
 of the mineral, ground as finely as possible in an agate 
 mortar, are treated with potassium chlorate and hydro- 
 chloric acid. After desiccation and extraction with hydro- 
 chloric acid and water, the insoluble substances may be 
 lead, calcium, barium, and strontium sulphates, silica, and 
 undecomposed mineral. By stirring well and filtering the 
 liquid whilst warm, the two former salts may generally, 
 however, be dissolved. The filtration should be performed 
 through a double filter, to prevent the pulverised mineral 
 from passing through. When the clear portion of the 
 solution has been poured upon the filter, add to the in- 
 soluble matter 5 c.c. of hydrochloric acid and 5 c.c. of 
 water ; then leave the mixture for two hours on the water- 
 bath at a boiling temperature, when, if care be taken to 
 stir well, the calcium sulphate will be completely dissolved. 
 Wash the insoluble portion in warm water, and pour it on 
 
40G ESTIMATION OF SULPHUR IN IRON ORES. 
 
 a filter, taking care to place below a flask specially to^ 
 receive that portion of mineral which may have passed 
 through the filter. The filtered liquid, whose volume is 
 about 50 c.c., should be rapidly boiled and mixed with 
 2 c.c. of a saturated solution of barium chloride. (This 
 amount is sufficient to precipitate the sulphuric acid 
 formed from 0*1 grm. of sulphur.) After cooling add to- 
 the mixture 5 c.c. of ammonia, sp. gr. 0-95, then stir well 
 with a glass rod, and leave the whole to rest at the ordi- 
 nary temperature for twenty-four hours. The clear solu- 
 tion should be decanted as completely as possible on to a 
 strong filter, and the precipitate stirred up with about 20- 
 c.c. of cold water, and then left to itself until it has quite 
 settled. If warm water is used without having added a 
 few drops of hydrochloric acid, a little oxide of iron will be 
 precipitated. The clear liquid is likewise thrown on to the 
 filter, and the operation is repeated several times with cold 
 water, and then two or three times with boiling water,, 
 without which precaution the barium sulphate will pass 
 through the filter. Finally collect the precipitate, and 
 wash it with warm water. The last drops of this water, 
 on being evaporated on a watch-glass, ought not to leave 
 more than a scarcely visible white ring. The precipitate 
 is then dried, heated to redness, and weighed. If it is 
 coloured with iron oxide it must be washed with a little 
 hydrochloric acid, dried in the water-bath, and taken up 
 with a few drops of acid and w r ater, and then the pre- 
 ceding operation repeated (w r ashing, drying, heating, and. 
 weighing). If the precipitate has only a feeble red colour, 
 which is often the case, this latter operation will be un- 
 necessary. 
 
 To dissolve the lead sulphate which may occur in the 
 insoluble portion, remove this from the filter with the end 
 of a feather, introduce it into a flask, and pour over it 
 10 c.c. of concentrated ammonium acetate. The solution, 
 after having been strongly agitated and heated in the' 
 water-bath, is carefully poured on to a filter. Then wash 
 the residue with a little ammonium acetate, and repeat 
 the treatment until a few c.c. of solution, acidulated with 
 
ESTIMATION OF SILICON IN IRON AND STEEL. 407 
 
 a little acetic or hydrochloric acid, is not rendered turbid 
 when warmed with barium chloride solution. The ni- 
 trate is then diluted, slightly acidified, and the sulphuric 
 acid precipitated by means of barium chloride. After the 
 lead sulphate has been removed, there may still occur 
 barium and strontium sulphates in the insoluble portion. 
 To decompose these salts the residues must be dried, 
 heated to redness, and weighed, and then fused with five 
 times their weight of pure dry sodium carbonate. The mass 
 is digested with water over a water-bath, the liquid filtered, 
 and the residue washed with warm water. The silicic acid 
 is separated from the solution, which contains the sodium 
 silicate, carbonate, and sulphate, by adding hydro- 
 chloric acid and drying on the water-bath. After filter- 
 ing, precipitate the solution with barium chloride. To 
 ascertain if the iron mineral contains gypsum or other 
 soluble sulphates, take 5 grammes, place them in 20 c.c, 
 of hydrochloric acid and 60 c.c. of distilled water, and 
 digest them for three hours on the water-bath, with 
 frequent agitation. The filtered solution is mixed with 
 barium chloride and 15 c.c. of ammonia ; then proceed as 
 already described. If there be present in the mineral 
 grains of iron- or copper-pyrites, or galena, they will only 
 give traces of sulphuric acid in this operation. 
 
 Estimation of Silicon in Iron and Steel. All who are 
 occupied in the analysis of iron and steel are aware how 
 very uncertain the estimation of silicon is when the 
 method hitherto used for its separation is followed, because 
 cast iron, and even bar iron and steel, are never found 
 absolutely free from intermingled slag. This slag is de- 
 composed by the ordinary method of dissolving the iron 
 in acids, and its silica then augments the amount of silica 
 formed from the silicon contained in the iron or steel* 
 This cannot be said of every sort of cast iron, but these 
 sometimes contain blast-furnace slag, although pig iron 
 containing slag may be considered as rare. It ought also 
 to be mentioned that crystallised silicon has been found 
 in crystallised cast iron from Krain, in the form of small 
 silvery plates, which are neither acted upon by boiling 
 
408 ESTIMATION OF SILICON IN IRON AND STEEL. 
 
 aqua regia nor by ignition in oxygen gas ; but they are 
 converted into silica by fusing with potassium and sodium 
 carbonates. 
 
 Crystallised silicon is insoluble in hot solutions of 
 sodium carbonate, but is soluble, with development of 
 hydrogen, in hot solutions of caustic potash, and also in 
 hot hydrofluoric acid. The accurate estimation of the 
 silicon in iron and steel has been effected by Dr. Eggertz, 
 who, after fruitless efforts to dissolve iron in highly diluted 
 organic or inorganic acids, which should have no effect on 
 the refinery slag, finally discovered such a solvent in 
 bromine, which, when mixed with water, dissolves the 
 iron without the slightest action on the accompanying slag. 
 But as experimenting with bromine in large quantities is 
 very disagreeable, trials were made to use iodine instead ; 
 and this, like bromine, has been proved to have no effect 
 on the slag, nor on iron oxide or proto-sesquioxide, or 
 manganese proto-sesquioxide. At the same time, bromine 
 dissolves iron quicker than iodine does, and is, perhaps, 
 more easily obtainable in the requisite state of purity. 
 Moreover, as continued experiments have shown that a 
 solution of sodium carbonate can separate finery slag from 
 the silica which has been formed by the use of iodine or 
 bromine on the silicon contained in the iron, the follow- 
 ing method for the estimation of silicon and slag in bar 
 iron or steel has been used and considered successful ; the 
 same method may be employed for cast iron, because 
 blast-furnace slag, when such is found, is not perceptibly 
 changed by iodine or bromine, nor by solutions of sodium 
 carbonate. Three grammes of bar iron or steel which 
 have passed through a sieve of 0*2 of a line are taken for 
 analysis. Fifteen grammes of iodine are added in small 
 portions at a time to 15 c.c. of water in a beaker of 100 
 c.c. capacity. The water must be previously boiled to 
 expel the air, which would otherwise oxidise the iron. 
 The iodine is stirred in the water with a glass rod, in order 
 to get rid of the air which has accompanied it, and the 
 floating iodine particles are allowed to sink. The beaker 
 with the iodine and water, which is kept covered with a 
 
ESTIMATION OP SILICON IN IRON AND STEEL. 409 
 
 watch- glass, is cooled in ice before the iron is put in, and 
 during the solution it is kept at the temperature of C. 
 For the first few hours it must be well stirred every hour, 
 or oftener, with a glass rod, but afterwards not so fre- 
 quently. In consequence of the low temperature and the 
 careful admixture of the iron (by which heat is prevented), 
 the solution may be performed without the least de- 
 velopment of gas, and the iron has less tendency to 
 become oxidised by the air when at this low tempera- 
 ture. By pressure, and by agitating with the glass 
 rod, the solution of the iron particles which collect 
 at the bottom of the beaker is much facilitated ; if no 
 lumps of the particles are visible, the beaker may be 
 kept at the ordinary temperature or, preferably, in ice 
 water. If some of the solution has dried on the sides of 
 the beaker or on the glass rod, it must be well moistened 
 with the same solution before water is added. About 30 
 c.c. of water, which should be very cold in order to pre- 
 vent the formation of basic salts, are added to the solution ; 
 it is then well stirred, left to settle, and the fluid with the 
 lighter particles of graphite is poured on to a filter of 2 
 in. diameter ; the filtration is kept up without interruption 
 until there remains only a somewhat heavy dark powder 
 of slag, &c., at the bottom of the beaker ; about 5 c.c. of 
 water, with a few drops of hydrochloric acid, are now 
 poured in and stirred with the glass rod ; if hydrogen is 
 given off, it is an indication that there is still some metallic 
 iron undissolved. The acidified water is quickly poured 
 on the filter, in order not to act on the slag. If a de- 
 velopment of gas is perceived, a little iodine, with sodium 
 carbonate and water, is added for the complete solution 
 of the iron, and the residue is thrown on the filter and 
 washed with cold water, until a drop of the filtrate gives 
 no reaction with a solution of 0-2 per cent, of potassium 
 ferrocyanide contained in a small porcelain crucible. 
 Solutions containing 0- 00001 gramme of ferric oxide per 
 c.c. show in this way very distinct reactions, particularly 
 if a drop of dilute nitric or hydrochloric acid be added. 
 The filtrate is evaporated to dryness, in which operation 
 
410 ESTIMATION OF SILICON IN IRON AND STEEL. 
 
 some of the iodine is sublimed away. Thirty c.c. of 
 hydrochloric acid, 1*12 sp. gr., are then added, and it 
 is again evaporated in order to obtain the silica which 
 may be dissolved in it. The filter, previously dried and 
 weighed, is again dried and weighed when containing the 
 precipitate. It is then ignited, and the residue weighed. 
 After ignition, the residue is boiled in a solution of soda, 
 in order to extract the silica, and weighed. It should be 
 observed that some part of the silica which has been 
 formed from the silicon in the iron may possibly unite 
 with the slag during the drying and ignition. In conse- 
 quence of this, it is difficult to extract it by means of a 
 soda solution, whence this method is not to be recom- 
 mended in exact estimations of silicon. 
 
 When using bromine as a solvent 6 c.c. must be taken 
 to 8 grammes of finely powdered iron or steel, with 60 c.c. 
 of water, which has been previously boiled and cooled to 
 C. ; and this temperature must be preserved by placing 
 the beaker in ice water until the solution is complete, 
 which usually takes place in two or three hours ; it is 
 cautiously stirred once or twice with a glass rod ; if stirred 
 hastily, the solution proceeds too violently. The further 
 operations are conducted in the same manner as when 
 using iodine. The bromine is most conveniently preserved 
 under water, and is taken up by a pipette, which is in- 
 troduced into the bottle, the upper end being kept closed 
 by the finger. 
 
 When it is preferred to dissolve iron or steel in lumps, 
 instead of in powder, this may be done ; but it is not then 
 necessary to place the beaker in ice water, as the metal is 
 less violently acted upon in this form. Several days are 
 required for the solution ; the iron, and particularly the 
 pieces of steel, must be occasionally cleaned from the 
 graphite which adheres to their surface. 
 
 In order to estimate the silica (formed from the silicon 
 in the iron) and slag, the filter, which contains graphite (in 
 combination with iodine or bromine and water), silica, and 
 slag, is unfolded, whilst it is still wet, on a watch-glass. 
 The contents are washed away from the filter (these should 
 
ESTIMATION OF SILICON IN IEON AND STEEL. 411 
 
 only occupy the lower half of the filter whilst in the 
 funnel) with a very fine jet from a wash-bottle (so as not 
 to obtain too much water) into a platinum or silver crucible 
 of the capacity of 30 c.c. The loosening of the mass may 
 be facilitated by a fine camel-hair brush. The water in 
 the crucible is evaporated to about 6 c.c., then mixed with 
 3 c.c. of a saturated solution of sodium carbonate free 
 from silica, and the crucible put in a water-bath, It is 
 kept in the boiling water 1 hour, during which time the 
 liquid is stirred two or three times, and the insoluble mass 
 crushed with a platinum spatula. The supernatant liquid 
 is carefully poured from the insoluble mass on to a small 
 filter, and to the mass in the crucible are added 1 c.c. of 
 saturated solution of sodium carbonate, and 2 c.c. of water. 
 When this has been boiled 1 hour, the whole contents of 
 the crucible are thrown on the filter and washed. The 
 solution of silica in soda is acidified by hydrochloric acid, 
 and mixed with the iron solution, and the whole evaporated 
 to dryness on a water-bath. When the solution attains" 
 the thickness of ordinary syrup, it is stirred very often 
 with the glass rod until the mass becomes a dry powder, 
 and heated until the smell of hydrochloric acid has nearly 
 gone off; the beaker is then placed in boiling water for 
 6 hours, 15 c.c. of hydrochloric acid of 1-12 sp. gr. are 
 then added, and the beaker left on the water-bath 1 hour* 
 As soon as the red powder is entirely dissolved 50 c.c. of 
 water are added ; and when no crystals of iron chloride 
 are visible, the solution is thrown on a filter and washed 
 with cold water, warm water forming basic iron salts which 
 make the silica appear red. The filter containing the silica 
 is dried and ignited in a porcelain crucible, gradually in- 
 creasing the temperature to a full red heat, and weighed ; 
 if the silica is coloured red by ferric oxide, a little hydro- 
 chloric acid, 1*19 sp. gr., must be poured into the crucible, 
 One decigramme of ignited and pure silica obtained from 
 analysis will dissolve in the above manner in 6 c.c. of a 
 saturated soda solution and 12 c.c. of water. If any 
 residue is observed after the second boiling, this arises 
 from some impurity which has united in small quantities 
 
412 ESTIMATION OF SILICON IN IKON AND STEEL. 
 
 with the silica, rendering it insoluble. When strong hydro- 
 chloric acid is boiled with this insoluble silica, it may 
 afterwards be dissolved. When the solution of silica is 
 diluted with water to the volume of 50 c.c. at the ordinary 
 temperature, it has no tendency to come into the form of 
 jelly. Quartz-powder is dissolved by the preceding 
 method, but very slightly, but ignited titanic acid and 
 finery slag are not acted upon, and the tersilicate slag from 
 blast-furnaces but very little. 
 
 When the silica is quickly exposed to a high tempera- 
 ture a considerable loss may arise from the spirting of the 
 water combined with the silica. Silica dried at 100 C. 
 has been proved to retain 1 equivalent of water to 3 
 equivalents of silica that is, about 6 per cent, of water, 
 which is lost by a strong ignition. The ferric oxide is 
 easily dissolved in the heat of a water-bath. The silica is 
 again thrown on a filter, w r ashed, dried, ignited, and 
 weighed. 
 
 To ascertain the purity of the silica, it may be mixed 
 in a platinum crucible with ten times its weight of pure 
 ammonium fluoride, diluted with water to the thickness of 
 syrup. The water must be evaporated on a water-bath, 
 and the crucible heated, with a cover on it, by a gradually 
 increasing heat over a spirit-lamp to a full red. If nothing 
 is left in the crucible, the silica was pure, and has passed 
 off as silicon fluoride ; but, if anything remains, the opera- 
 tion with ammonium fluoride must be repeated until a 
 constant weight is obtained. When iron contains tungsten, 
 some tungstic acid is formed, and this accompanies the 
 silica for the most part, being dissolved by the soda solu- 
 tion, but it is not volatilised by the use of ammonium 
 fluoride. Yanadic acid also accompanies the silica, behav- 
 ing as tungstic acid. Instead of using ammonium fluoride, 
 it is preferable to use hydrofluoric acid, with which the 
 silica is moistened, and the evaporation is conducted on a 
 water-bath. The mass left on the filter from the soda 
 solution may be composed of besides graphite slag, iron 
 oxide, titanium oxide, &c. (but not copper, at least when 
 the iron does not contain more than 1 per cent.); this is 
 
ME. TURNERS PROCESS. 413 
 
 dried, ignited, and weighed. The method of separating 
 iron oxide and slag, when the iron or steel contains both 
 these, is not yet known. If the composition of slag were 
 always alike (which it is not) it would be easy to calculate 
 its amount from either the silica or iron oxide obtained in 
 the analysis. In a piece of very red-short Bessemer iron 
 which contained no sulphur, by several experiments 0'3 
 per cent, of iron oxide has been obtained, and only traces 
 of silicon. After ignition, the iron oxide may possibly be 
 found as sesquioxide. The amount of oxygen, in case the 
 redshortness is due to this, as it probably is, amounts to 
 less than Ol per cent. 
 
 When the iron or steel for analysis contains titanium, a 
 part of this substance follows the slag in the form of titanic 
 acid. If such is the case, this must be melted with ten 
 times its weight of acid potassium sulphate, by which it 
 is dissolved ; the mass is dissolved in cold water, and the 
 solution precipitated by boiling ; the weight is determined, 
 and subtracted from that of the slag. 
 
 Mr. Thomas Turner estimates silicon in iron and steel 
 in the following way : 
 
 For a sample of cast-iron, 2 grammes of metal in the 
 form of borings are placed in a clean beaker of about 8 
 ounces capacity, and covered with about 50 c.c. of water; 
 5 c.c. of sulphuric acid are then added, and the beaker 
 kept covered while solution proceeds. The liquid is then 
 evaporated till quite solid, after which the residue is ex- 
 tracted with 100 c.c. of boiling water. The insoluble portion 
 is washed with boiling hydrochloric or nitric acid, and 
 afterwards with water so long as any iron is extracted. It 
 is necessary in all cases to test the filtrate by potassium 
 ferrocyanide or thiocyanate. The insoluble portion is dried, 
 ignited, and weighed in the usual way. 
 
 In the analysis of steels containing only a small quan- 
 tity of silicon, results by this method agree fairly well 
 with those obtained by other trustworthy processes. In pig- 
 iron of specially good quality good results are also obtained. 
 
 The conclusions arrived at from Mr. Turner's experi- 
 ments are as follows : 
 
414 BASIC CINDER IN IRON. 
 
 1. That with cast irons of specially good quality the 
 silicon can be correctly estimated by evaporation with 
 dilute sulphuric acid. 
 
 2. With phosphoric irons the residue obtained, though 
 white, is often impure, and should be further treated in 
 order to obtain accurate results. 
 
 3. "With phosphoric irons containing titanium, the 
 silica is contaminated not only with iron, but also with 
 titanic oxide and phosphoric acid. The residue may be 
 very nearly white and still contain 20 per cent, of sub- 
 stances other than silica. 
 
 4. On treatment with aqua regia the colour of the 
 residue is usually an indication of its purity. 
 
 The amount of silicon in grey charcoal pig-iron is about 
 2*7 per cent, and in spiegeleisen 0-8. per cent. The 
 amount of silicon in pig from coke blast-furnaces is rarely 
 more than 4 per cent. The least quantity of silicon in 
 grey cast iron is about 0*2 per cent., and in white (spiegel- 
 eisen) O'Ol per cent. The amount is usually from 1 to 2 
 per cent, in cast iron suitable for the Bessemer process, 
 and in pig-iron for puddling about 0'5 per cent. The 
 amount of silicon in iron of different degrees of hardness 
 from the same charge of the blast-furnace ought to be 
 pretty well judged by the fracture, after some estimations 
 have been made by analysis. 
 
 Estimation of Basic Cinder and Oxides in Manufac- 
 tured Iron. The principal value attached to estima- 
 tions of slag and oxides in defective iron is the almost 
 absolute certainty of discovering whether to the chemical 
 constituents of the metal or to its careless manufacture 
 may be attributed the faults observed. When we con- 
 trast the ' life ' of an iron rail of fair chemical purity with 
 ' mild ' steel rails of nearly the same composition, it is 
 forcibly suggested to even the most superficial observer, 
 that the duration of a rail is proportionate to the cohesion 
 of its metallic particles, due principally to freedom from 
 mechanical impurities. Engineers have ever been alive 
 to the necessity of suitable machinery for expressing the 
 fluid cinder from the iron as it comes from the puddling 
 
BASIC CINDER IN IRON. 415 
 
 and re-heating furnaces ; but how rarely is the estima- 
 tion of cinder asked for at the hands of the analyst! We 
 occasionally find extremely pure irons and steels pure 
 at least by the result of routine analysis which are con- 
 demned when subjected to mechanical tests. In some 
 cases a crystalline structure of the metal may weaken the 
 iron ; but it is usually the cinder which, by breaking the 
 lines of continuity in molecular structure of the metal, 
 determines its fracture. 
 
 We regularly estimate the ordinary chemical consti- 
 tuents of a manufactured iron : why not its contained 
 oxides of manganese and iron, and basic cinder, which, 
 when present in excess, weaken the iron and so detract 
 from its value ? 
 
 The cause of the omission seems to be that no process 
 with the exception of Eggertz's is suitable for technical 
 work. Fresenius's process is, without doubt, accurate ; 
 but what process can be admitted into an industrial labo- 
 ratory (where a complete analysis is required in two days 
 or less), which for each experiment requires a rod 
 of the metal and several days for the completion of the 
 analysis ? 
 
 An estimation of cinder, &c., by the following me- 
 thod, devised by Mr. Bettel, only occupies an hour or so, 
 and has the advantage of yielding constant results : Five 
 grammes of the borings are heated with a solution of 10 
 c.c. bromine and 35 grammes potassium bromide in 150 
 c.c. water. Heat is continued until the iron is dissolved, 
 the solution filtered through a 4-inch Swedish paper (pre- 
 viously washed with hydrochloric acid and boiling water), 
 drained, and washed with a solution of sulphurous acid 
 containing 5 per cent, hydrochloric acid. When the 
 filtrate is practically free from iron, wash with boiling 
 water containing \ per cent, hydrochloric acid ; then with 
 pure water rinse into small platinum dish ; evaporate to low 
 bulk, and dissolve out silica by means of hot solution of 
 carbonate of soda. Boil, dilute, filter, wash with hot 
 water ; then with a ^-per-cent. solution of hydrochloric 
 acid ; finally with water. Dry, ignite, and weigh. Esti- 
 
416 BASIC CINDER IN IRON. 
 
 mate the silica in the residue in the ordinary manner. 
 The bromine in the first filtrate may be recovered as 
 bromide of potassium in an obvious way. 
 
 Some analysts object to the bromine process on account 
 of the vapour contaminating the air of the laboratory. 
 To those Mr. Bettel recommends the following process, 
 which has given good insults : Five grammes of the iron 
 rather finely divided are dissolved in 60 c.c. clear solution 
 of cupric chloride (1 in 2) mixed with 100 c.c. saturated 
 solution of potassium chloride. When no particles of 
 iron can be felt by the aid of a glass rod, add 50 c.c. of 
 dilute hydrochloric acid (1 in 20) ; boil and filter through 
 a 4-inch Swedish paper previously moistened with hot 
 hydrochloric acid (1 in 3), then with saturated solution of 
 potassium chloride. Wash the residue on the filter with 
 potassium chloride solution till all the copper is removed, 
 then with hot dilute hydrochloric acid (1 in 50), finally 
 with hot water. Separate the silica as before, ignite and 
 weigh. If the copper obstinately adheres to the paper, as 
 sometimes happens, slip over the tube of the funnel a piece 
 of india-rubber tubing with clip (or plugged with glass 
 rod) ; fill up funnel with strong liquid ammonia ; cover, and 
 allow to remain for half an hour ; then proceed with dilute 
 acid, &c., as before. 
 
 The results of both processes agree with Fresenius's 
 galvanic method. 
 
 Estimation of Phosphorus in Iron and Steel. The im- 
 portance of ascertaining the quantity of phosphorus in 
 iron is very great ; for although the presence of a very 
 small quantity of phosphorus in cast iron does not produce 
 any sensible modification in its proprieties, it nevertheless 
 loses its most essential qualities when the proportion of 
 phosphorus amounts to some thousandths. Almost all the 
 methods hitherto proposed consist in treating cast iron 
 with oxidising agents in such a manner as to cause the 
 phosphorus to pass into the condition of phosphoric acid, 
 which is estimated as a magnesian compound. Several 
 sources of error are inherent to this method, to avoid 
 which M. Tantin liberates the phosphorus as an hydrogen 
 
ESTIMATION OF PHOSFHOKUS IX IRON AND STEEL. 417 
 
 compound. Experiment shows that there is not the least 
 trace of phosphorus in the residue after the complete at- 
 tack of the cast iron by hydrochloric acid, which fact is not 
 surprising if it be considered what strong affinities phos- 
 phorus has for hydrogen. The hydrogen phosphide pro- 
 duced by the action of hydrochloric acid upon cast iron 
 is almost always accompanied by sulphuretted, arseniu- 
 retted, and carburetted hydrogen. In order to effect the 
 separation of these gases, first pass them into a WoulfFs 
 flask containing a solution of potash, which absorbs the 
 sulphuretted hydrogen ; the other gases are then made to 
 traverse a solution of silver nitrate, which transforms the 
 arseniuretted hydrogen into silver arsenite, soluble in the 
 slightly acid liquid, whilst the phosphuretted hydrogen 
 precipitates the silver solution and forms an insoluble phos- 
 phide. The phosphorus being thus completely separated 
 from the sulphur and arsenic, its estimation is effected in 
 a simple manner. The phosphide of silver is treated with 
 aqua regia, and thus transformed into silver chloride and 
 phosphoric acid, which is precipitated in the state of am- 
 monia-magnesian phosphate. 
 
 Mr. J. B. Mackintosh has conducted several careful 
 experiments on M. Tantin's method, and finds that to 
 secure accuracy the steps to be followed are : 
 
 1. Solution in hydrochloric acid in a stream of oxygen 
 or air, absorbing the escaping gases in permanganate acidi- 
 fied with sulphuric acid. 
 
 2. Heating the solution to boiling, stopping the passage 
 of the oxygen current, and carefully adding an excess of 
 sulphurous acid solution, and continuing to boil till the 
 precipitated manganese binoxide in the absorption-flask 
 is re-dissolved. This boiling should last several minutes 
 to insure the completion of the reaction. 
 
 3. Disconnecting the junctions between the absorption- 
 flasks and solution-flask, and between the solution-flask 
 and the acid-flask, and allowing to cool. 
 
 4. Mixing solutions, filtering out residue which is 
 placed (with filter-paper) in a porcelain dish, oxidising 
 
 E E 
 
418 ESTIMATION OF PHOSPHORUS IN IRON AND STEEL. 
 
 with nitric acid and potassium chlorate, and evaporating 
 to dryness.* 
 
 5. Boiling the solution till the excess of sulphurous 
 acid is expelled, adding a few c.c. of permanganate to per-, 
 oxidise a little of the iron, and precipitating basic acetates. 
 Boiling the nitrate for other precipitates, to insure getting 
 all the phosphoric acid present. 
 
 6. Dissolving these precipitates in hydrochloric acid 
 and adding to the solution of the residue, in which by this 
 time the paper will have been thoroughly destroyed. 
 
 7. Evaporating to dryness for silica, and proceeding as 
 usual with the molybdate precipitation. 
 
 The form of apparatus used in these estimations is 
 shown in fig. 104. It consists of a flask, A, to hold the acid ; 
 
 FIG. 104. 
 
 a flask, B, for the iron ; and flasks c and D, of which c is 
 generally left empty to catch condensed steam, and the 
 two flasks D contain the absorbent liquid. 
 
 The absorbent liquid in D is a solution of potassium 
 permanganate acidified with nitric acid. The iron having 
 been placed in B, and the necessary amount of dilute 
 hydrochloric acid to dissolve it in A, the whole apparatus 
 is filled with oxygen, a current of which is kept passing 
 through the whole time. When the air is all expelled 
 
 * It is better to filter out the residue instead of filtering it out with the 
 first basic acetate precipitate, because it is difficult to work accurately on a 
 muddy liquid, and because there is a probability of loss of phosphorus by the 
 continued action of the hot acid solution on the residue after the sulphurous 
 acid has been expelled. By proceeding as directed this chance of loss is 
 removed. 
 
ESTIMATION OF PHOSPHORUS IN IRON AND STEEL. 419 
 
 the flask A is inverted above the level of B, so that its 
 contents flow into B and the solution of the iron com- 
 mences. The solution is heated to boiling for some 
 minutes ; and when all has dissolved with the exception 
 of the insoluble portion, the apparatus is disconnected, 
 the insoluble residue filtered from the solution, the solution 
 -oxidised with nitric acid and treated as described above. 
 
 The saving of time by this process is the greater as 
 the percentage to be estimated is less. In a steel of low 
 percentage, where it is necessary to use 10 grms. for the 
 -estimation, necessitating the use of 120 c.c. of nitric 
 acid for solution, by the old method, the time taken for 
 the evaporation of this amount of liquid, for the subse- 
 quent thorough drying of the residue, and for the reso- 
 lution and reduction of the iron to the ferrous state, is 
 evidently much greater, and the operations followed are 
 much more tedious, than by this method, where the phos- 
 phorus is concentrated at the start in but a few milli- 
 grammes of iron in very little bulk of solution, enabling the 
 subsequent evaporation to dryness and thorough drying 
 of the residue to be performed in a very short period. 
 
 In M. Tantin's experiments he used cast iron only. Mr. 
 Mackintosh's have been conducted on pig and wrought 
 iron. It is possible that the difference in the results may 
 be partially accounted for by different modes of occurrence 
 of phosphorus similar to the different forms of carbon ; but 
 this is only offered as a suggestion. 
 
 In calculating the percentage of phosphorus from the 
 weight of the yellow precipitate, Mr. Mackintosh uses in 
 all cases the figure of 1'63 per cent, phosphorus in the 
 precipitate. 
 
 Dr. J. Lawrence Smith was engaged off and on for two 
 or three years, examining the question of the estimation 
 of phosphorus in iron and steel, making several hundreds 
 of variously modified experiments, and repeating the 
 details of processes adopted by various chemists. The 
 following method was ultimately adopted as affording the 
 most speedy and accurate results. 
 
 Quantity of Iron employed. It is customary to employ 
 
 E E 2 
 
420 ESTIMATION OF PHOSPHORUS IN IRON AND STEEL. 
 
 1 grm. for pig iron, and 2 to 3 grms. for malleable iron 
 and steel ; but Dr. J. L. Smith employs but 1 grm. for all 
 varieties of iron ; for even where the iron or steel contains 
 one- thousandth and less of phosphorus, as satisfactory 
 results are obtained as where 2 and 3 grms are employed. 
 
 Solution. The iron, say 1 grm., is placed in a por- 
 celain capsule of about 100 or 150 c.c., and 3 or 4 
 c.c. of water added ; the capsule is placed on a water- 
 bath, and 10 to 15 c.c. of aqua regia are added little by 
 little ; the aqua regia is prepared in advance in the usual 
 way with 2 parts hydrochloric acid and 1 part nitric acid. 
 The contents of the capsule are now evaporated to dry- 
 ness over the water-bath, or more speedily on an iron 
 plate ; the capsule with its contents is then placed in an 
 air-bath and heated from 140 to 150 C. for from 30 
 minutes to 1 hour, thus rendering all the silica insoluble ; 
 3 or 4 c.c. of hydrochloric acid with an equal quantity of 
 water are added to the dry residue, and then warmed 
 gently over a water-bath or lamp ; the iron is re-dissolved, 
 a little more water added, the solution filtered with the 
 filter-pump, the filtrate placed on a narrow graduated 
 measure of 100 c.c. capacity, and sufficient water added 
 to make the liquid contents 100 c.c. ; the whole is well 
 shaken to make the solution uniform. The next step is 
 to concentrate all the phosphorus into a limited amount of 
 the iron. 
 
 Concentration of the Phosphorus. From 90 to 92 c.c. 
 of the last solution is placed in a capsule of 300 or 400 
 c.c. capacity, either of porcelain or platinum the latter 
 by preference and 100 c.c. of water added ; the iron 
 oxide is now reduced to iron protoxide by ammonium 
 sulphite.* Two or three centimetres of the ammonium 
 sulphite are added to the iron solution and the contents of 
 the capsule are boiled until all the sulphurous acid is driven 
 off, this stage of the process being recognised by the sense 
 of smell. By putting a small drop of the solution on the 
 
 * Equal parts of ammonia and water are placed in a bottle and an excess 
 of sulphuric acid passed through ; the operation lasts for several hours, using 
 a mixture of charcoal and sulphuric acid. Once prepared, it keeps very well, 
 when kept from the light. 
 
J. LAWRENCE SMITH'S PROCESS. 421 
 
 end of a glass stirrer into a weak ammonia solution we 
 readily recognise the complete conversion of the oxide, for 
 the precipitate is nearly white. Of course during the 
 whole of the above process the solution is acid, with the 
 excess of hydrochloric acid. Ammonia is now added 
 slowly to the warm solution until a little of the greenish 
 precipitate remains undissolved ; about 20 c.c. of acetic 
 acid is now added to the solution (which immediately re- 
 dissolves the precipitate), and then 1 or 2 c.c. of ammonia 
 acetate solution; finally, the 8 or 10 c.c. of original solu- 
 tion remaining in the graduated glass is added with 200 
 or 300 c.c. of water. 
 
 The whole contents of the large capsule is boiled 
 gently from one-half to one hour, and if necessary the 
 water renewed as it is evaporated. The result is the for- 
 mation of a basic per-salt of iron containing practically 
 all the phosphorus that was originally in the gramme of 
 iron used. 
 
 Separation of the Phosphorus from the above Precipi- 
 tate. With a filter-pump on a 3^-inch filter, the last pre- 
 cipitate is collected in 15 or 20 minutes. The precipitate 
 is not washed, but a mixture of 5 or 6 c.c. of hydrochloric 
 acid, with an equal quantity of water, is warmed in the 
 capsule in which the boiling has taken place, so as to 
 dissolve the adhering oxide of iron ; the hot acid solution 
 is thrown on the filter in the funnel, detached from the 
 pump ; the filtrate is readily dissolved, and passes into some 
 convenient vessel, and the filter washed once or twice ; 
 this solution is placed in a porcelain capsule and evapo- 
 rated to dryness over a water bath or on a hot plate. The 
 former is preferable, although it takes a longer time. To 
 the dry, but not over-heated residue is added 1 to 2 c.c. 
 of nitric acid, with an equal quantity of water. This will 
 furnish a clear solution if there be no titanium in the 
 iron ; if the latter be present, there will be formed a floccu- 
 lent precipitate that can be readily separated by a filter 
 prior to the last treatment. 
 
 The last Treatment. The solution now need not be 
 more than 10 or 20 c.c., to which ammonia is to be added 
 
422 ESTIMATION OF PHOSPHORUS IN IRON AND STEEL. 
 
 until the precipitate first formed is no longer re-dissolved ; 
 then add a few drops of nitric acid to clear up the solution 
 completely, in which the phosphorus is supposed to have 
 been concentrated. 30 c.c. of molybdic acid solution is 
 now added to the last solution in a small beaker, which is 
 then warmed for 15 or 20 minutes to a temperature of 
 80 0., and agitated with a glass rod. The phosphorus is 
 precipitated as the double ammonia-salt, and settles as a 
 chrome-yellow powder in less than 30 minutes, and is 
 ready for collection on a double filter ; * although it is 
 better to allow two hours or more time to elapse before 
 filtering and washing with the filter-pump. As the filter 
 is very small, it is readily washed with a little distilled 
 water. 
 
 After washing, the double filter is placed in an air-bath 
 heated to about 120 C., and in about 30 minutes weighed 
 by separating the filters ; the complete dryness is verified 
 by a second heating in the air-bath. 
 
 Of the phospho-molybdate every ] 00 m.g. will contain 
 1-63 m.g. of phosphorus, or 3*74 m.g. of phosphoric acid. 
 The result of this method of analysis will indicate a very 
 minute quantity of phosphorus less than what is contained 
 in the iron, but so small as not to affect the practical result, 
 and will be more accurate, certain, and speedy than if 
 estimated as magnesian phosphate. 
 
 Estimation of Manganese in Iron. After the silicon 
 is estimated in the iron or steel by Eggertz's method 
 (p. 408), the manganese may be estimated in the same- 
 amount of material. The filtrate from the silicate is diluted 
 with water until it measures 400 c.c., and accurately 
 divided into two portions of 200 c.c. each, one of which is 
 set on one side, and in the other the manganese is estimated 
 in the following manner : (In the case of wrought iron 
 
 * When filtering a precipitate to be weighed on the filter, a double filter 
 is used, each of the same size ; they are weighed one against the other and 
 exactly balanced by the weights ; on the lighter one a -f mark is put with 
 pencil, and the number of m.g. that it is lighter than the other. As only a 2 
 or 3 inch filter is used, the difference in weight between the filters does not 
 usually exceed 10 or 20 m.g. A number of these double filters (with the differ- 
 ence marked on them) may conveniently be kept ready for this or any other 
 purpose. 
 
ESTIMATION OF MANGANESE IN IRON. 423 
 
 and steel, where 3 grammes are taken, the solution is 
 diluted to 200 c.c., and the manganese estimated without 
 dividing the solution.) A saturated solution of sodium 
 carbonate is added to the solution until it becomes nearly 
 neutralised, appearing of a deep brown colour ; water 
 containing 5 per cent, of sodium carbonate is then added, 
 drop by drop, until a slight turbidity occurs in the solu- 
 tion ; and if, on standing in the cold, this turbidity rather 
 increases than diminishes, sufficient carbonate has been 
 added (if too much sodium carbonate has been added, and 
 a precipitate is thrown down, the excess must be neutralised 
 by hydrochloric acid) ; to the slightly turbid solution add 
 1-^ c.c. of hydrochloric acid, and heat on the water-bath 
 until the solution becomes clear ; dilute with about half 
 as much water as the volume of the solution, and add 30 
 c.c. of a saturated solution of sodium acetate ; boil for a 
 quarter of an hour ; allow the precipitated iron to settle, 
 and decant the clear liquid through a filter, washing the 
 iron by decantation with boiling water containing -J per 
 cent, of sodium acetate ; finally, throw the iron o$ the 
 filter, and continue the washing until bromine water gives 
 no reaction, showing that all the manganese has passed 
 through the filter ; evaporate the filtrate down to 400 or 
 500 c.c. ; and at the temperature of 50 C. add a few drops 
 of bromine to precipitate the manganese, and keep it near 
 to that temperature for twelve hours, stirring occasionally 
 with a glass rod. The solution after the addition of the 
 bromine becomes of a yellow or brownish colour, but 
 should be perfectly colourless before filtering. The man- 
 ganese is now thrown on a filter which has been dried at 
 100 C., and accurately weighed, washed with cold water 
 containing 1 per cent, of hydrochloric acid, dried at 100 
 C., and weighed. The precipitate is a hydrated manga- 
 nese oxide, containing 59'21 per cent, of manganese. The 
 .precipitate may also be ignited in a porcelain crucible at 
 a white heat, and is then an anhydrous manganese oxide, 
 containing 72-05 per cent, of manganese. If copper is 
 present it must be removed previous to the precipitation 
 of the manganese ; or the amount of copper found in the 
 
424 ESTIMATION OF MANGANESE IN IRON. 
 
 ignited oxide, and then an equivalent amount of copper 
 oxide, subtracted from the total weight of the precipitate. 
 
 In using the first method, 20 grains of the finely 
 divided spiegeleisen are completely dissolved in hydro- 
 chloric acid, diluted, and a current of sulphuretted 
 hydrogen passed through the liquid. After standing for 
 twelve hours, the solution is filtered and washed with 
 water containing sulphuretted hydrogen ; the filtrate is 
 boiled, 10 grs. of potassium chlorate added, the iron sepa- 
 rated, and the manganese estimated in the usual manner. 
 If the method used be that of estimating the copper in 
 the precipitate, the estimation must be made with the 
 greatest care, on account of the small quantity of copper 
 present ; the solution must be decanted immediately the 
 zinc is completely dissolved, and excess of acid must be 
 carefully avoided ; otherwise the film of copper will par- 
 tially re-dissolve. It is evident that if the precipitation 
 be effected by ammonium sulphide or sodium carbonate 
 separation or estimation of the copper is likewise neces- 
 sary. 
 
 Mr. E. Eiley ('Chemical News,' April 27, 1877) gives 
 the following instructions on the estimation of manganese 
 in spiegeleisen, and in many auriferous iron ores : 
 
 There are two methods now in use (a) The Direct 
 Method. The pulverised spiegeleisen (about 1 grm.) is 
 dissolved in dilute nitric acid, sp. gr. 1*2, a little potassium 
 chlorate and hydrochloric acid added to destroy the soluble 
 organic matter from the combined carbon ; the solution, 
 diluted to about a litre, is neutralised with sodium or ammo- 
 nium carbonate, sodium or ammonium acetate added, the 
 solution boiled, the basic iron peracetate allowed to settle, 
 and filtered off. This precipitate is re-dissolved in hydro- 
 chloric acid, and the process repeated to insure complete 
 separation of the manganese. The filtrates are evaporated 
 to 1-^ litre, allowed to cool, 2 to 4 c.c. bromine added, 
 the solution well shaken, 0*88 ammonia added in excess, 
 the solution heated gradually for an hour, boiled for a 
 few minutes, the precipitate allowed to settle, filtered (the 
 filtrate should be evaporated and tested for manganese), 
 
MR. EILEY'S PROCESS. 425 
 
 dried, and ignited in a muffle, or over a gas blowpipe for 
 half an hour. 
 
 (b) The Indirect Method. The finely powdered spiegel- 
 eisen (about 1 grin.) is dissolved in dilute sulphuric or in 
 hydrochloric acid, the liquid diluted with recently boiled 
 and cooled distilled water, and the iron estimated volu- 
 metrically ; to the percentage of iron thus obtained 5 
 per cent, is added for carbon and impurities ; the difference 
 is assumed to be manganese. 
 
 The results obtained by this method are usually too low, 
 from the formation of soluble organic matter during the 
 process of solution. This error can be obviated by using 
 nitric acid for a solvent, evaporating to dryness and heat- 
 ing ; the iron and manganese oxides are then dissolved in 
 hydrochloric acid, the solution largely diluted, and reduced 
 with sodium sulphite. The results thus obtained agree 
 very closely with the direct method. Thus, for all practical 
 purposes, the indirect method is sufficiently accurate, and 
 can be accomplished in one hour, the direct method re- 
 quiring five or six hours. The author strongly recom- 
 mends the use of ammonium acetate and carbonate, instead 
 of the corresponding soda salts in the direct method ; and 
 proves by check experiments with pure manganese sul- 
 phate, &c., the statements of Fresenius and others, that the 
 presence of ammoniacal salts prevents the complete pre- 
 cipitation of manganese by bromine and ammonia, to be 
 erroneous. On the other hand, if soda salts be used, the 
 ignited precipitate will contain soda. The author con- 
 siders that sulphur cannot exist in spiegeleisen. He es- 
 timates the carbon by dissolving the iron in neutral 
 copper chloride, and after complete solution of the iron 
 and precipitated copper, the carbon is filtered on asbestos, 
 and burnt with copper oxide in a current of oxygen. 
 The carbon estimations by the colour test are unsatis- 
 factory for high percentages of carbon. According to the 
 author, the percentage of carbon varies with the percentage 
 of manganese. 
 
426 ASSAY OF MANGANESE IN SPIEGELEISEN. 
 
 Assay of Manganese in Spiegeleisen, Ferro-manganese, and 
 the most important ores. 
 
 W. Kalman and Alais Smolka have ascertained that 
 manganous oxide when opened up with a flux of borax 
 and potassium-sodium carbonate with access of air yields 
 an oxidation product containing 5 atoms available oxy- 
 gen and 6 atoms manganese. The flux is obtained by 
 melting in a platinum capsule 2 parts borax glass and 
 3 parts of the double carbonate, and pulverising the 
 very hygroscopic mass while still warm. There is be- 
 sides required a solution of ferrous sulphate and a per- 
 manganate solution. [The former is obtained by dissolving 
 about 100 grammes ferrous sulphate in 1,000 c.c. water, 
 acidulating with sulphuric acid, filtering, and mixing the 
 solution with 100 c.c. of pure undiluted sulphuric acid. 
 The permanganate solution is standardised for iron, 1 
 c.c. of the solution preferably representing 0-0025 grm. 
 iron. Its standard for manganese can be calculated from 
 the following proportion : iron standard : x 10-56 : 
 6-55. 
 
 For applying the method, from 0-15 to 0*30 grm. 
 of the sample, very finely ground, is ignited for fifteen 
 minutes in an open platinum crucible with a Bunsen 
 burner, and then more strongly with a blast. By the 
 ignition the manganese is chiefly converted into mangano- 
 manganic oxide. The crucible is let cool, covered, and 
 about 20 parts of the flux weighed in. Heat is slowly 
 applied till the mixture is melted, care being taken that 
 not much spirts up upon the lid, as such portions become 
 oxidised to manganate and make the result too high. 
 The contents of the crucible are kept in a state of fusion 
 for fifteen to twenty minutes, the lid is then removed, the 
 crucible placed slanting, and the fusion is continued for 
 five minutes longer, stirring with a platinum spoon. 
 
 Equal quantities (10 to 15 c.c.) of the iron solution 
 are poured into two beakers and diluted. The contents 
 of the crucible are dissolved in the solution in one of the 
 
ASSAY OF MANOANESE IN SPIEGELEISEN. 427 
 
 beakers, adding a little strong sulphuric acid if requisite, 
 when the manganese compound formed oxidises a part 
 of the iron. The solution of ferrous sulphate in both 
 glasses is then titrated with permanganate. The difference 
 multiplied by the manganese standard gives the propor- 
 tion of manganese in the sample. In cases where the 
 manganese exists as a silicate the results are only approxi- 
 mate. It is suitable for all cases where the manganese in 
 the sample may be converted into mangano-manganic 
 oxide by simple ignition, where the proportion of man- 
 ganese is at least 1 to 2 per cent., and where no other 
 substance is present which can become capable of giving 
 up oxygen to a ferrous solution, such as chlorine. 
 
 The accurate estimation of manganese in spiegel- 
 eisen is of commercial importance ; as, being the most 
 important constituent, the value of the material is fre- 
 quently judged by the percentage of that element alone, 
 while the error introduced by the presence of copper is 
 aggravated by the fact that not only is copper worthless 
 but absolutely injurious. 
 
 The following method for the assay of manganese in 
 iron and steel by Mr. Samuel Peters, Bay State Iron Works, 
 South Boston, is not new in principle, but has given very 
 satisfactory results : 
 
 Dissolve O'l grm. pig iron or steel in 3 or 4 c.c. 
 nitric acid, about 1-2 sp. gr., and boil gently in a long 
 test-tube (about 8 inches long and f-inch diameter) for 
 five or ten minutes, or until solution is complete ; then 
 add an excess of plumbic oxide, say 0-2 or O3 grm., 
 and boil again two or three minutes.* Cool the tube and 
 its contents in water. Filter through asbestos, washing 
 out the test-tube and the residue on the filter with dis- 
 tilled water until all the colour has been washed through. 
 Transfer to a graduated tube (f-inch in diameter), hold- 
 ing 50 or 60 c.c., graduated in 0*2 c.c., and compare with 
 a standard solution of permanganate held in a tube for 
 that purpose. The comparison is made in the same 
 
 * It is unnecessary to filter off graphite in pig iron before boiling with 
 plumbic oxide. 
 
428 ASSAY OF MANGANESE IN SPIEGELEISEN. 
 
 manner as that in the Eggertz method when estimating 
 combined carbon in steel, &c. The solution under com- 
 parison is then diluted and well mixed with distilled 
 water (by pouring the contents of the graduated tube 
 into a small dish, and then transferring to the tube again), 
 until its colour is exactly of the same intensity as the 
 standard solution. Having attained to this point, the 
 number of c.c. is noted, and the result is obtained by 
 multiplying each c.c. by O'OOOl. Each c.c. is equivalent 
 to 01 per cent, manganese when Ol grm. of iron is 
 taken for analysis. 
 
 For irons containing O'lO to 0*35 per cent, manganese, 
 Ol grm. is the proper quantity ; but if there be, say, 
 0-8 to TOO per cent., it is best to take 0-1 grm. and 
 divide the solution (before adding the lead peroxide) in 
 four equal parts, and use O25 for the estimation, taking 
 another 0'25 for a second estimation. In case of a high 
 percentage, as 1 per cent., if O'l grm. is taken the re- 
 sults are too low on account of some of the manganese 
 escaping oxidation. This agrees with the observations 
 of others. With an unknown iron, one or two trials 
 with O'l grm., or half that quantity, will point out the 
 probable amount, and so be a guide for the next trial. 
 If the amount of iron taken does not yield more colour 
 than corresponds to 25 to 35 c.c. of standard hue, it may 
 be safely said that all the manganese is oxidised. It is 
 as well to take this volume as the guide to the quantity 
 of iron to be taken. The quantity of manganese in the 
 liquor to be tested should not exceed 0'4 of a millgramme, 
 and certainly not over half a milligramme. By taking 0*1 
 grm. of a spiegeleisen containing nearly 12 per cent, 
 manganese, and diluting to 50 c.c., and taking 2 c.c. or 
 0'04 for the estimation of the manganese, very nearly the 
 proper amount of manganese is obtained. This seems to 
 show that if the division of the solution can be accurately 
 made, and the bulk of the coloured liquid can be kept 
 down well, the amount of manganese in spiegeleisen can 
 be estimated very fairly. 
 
 Combined carbon in large quantity does not inter- 
 
ASSAY OF MANGANESE IN SPIEGELEISEN. 429 
 
 fere with the accuracy of the method, for a steel con- 
 taining 2 per cent, combined carbon and only 0-8 per 
 cent, manganese was found to give good results by this 
 method. 
 
 The standard is made by diluting a permanganate of 
 potash solution of known strength until each c.c. = 0*00001 
 grm. manganese. 
 
 For example, a T n F solution will contain 3-16 grm. 
 permanganate in 1000 c.c. or 0*0011 grm. manganese 
 per c.c. ; if this be diluted 110 times it will give the 
 required strength. The standard is contained in a tube 
 of the same bore as the one used for the analysis ; or else 
 the standard is put in the latter one, and a solution of 
 permanganate put into a tube of nearly the same bore, 
 and diluted until it exactly corresponds with the standard 
 solution, when it will serve as a standard. 
 
 Permanganic acid of the proper hue keeps better than 
 permanganate of potash of the same hue, and is of course 
 easily made by adding nitric acid to the latter. The time 
 occupied in obtaining a result by this method is very 
 short (about half an hour), and it is a method that will 
 prove of advantage in analysing steel made by the Bes- 
 semer and Siemens-Martin processes. 
 
 Mr. Galbraith (< Chemical News,' February 4, 1876) 
 has given the following simple and accurate process for 
 the assay of manganese in spiegeleisen : 1 gramme of the 
 spiegeleisen is dissolved in nitric acid (sp. gr. 1-20) in a 
 small round-bottomed flask, and evaporated to dryness. 
 When dry, the flame, which may be either a spirit-lamp or 
 a Bunsen burner, is turned so that the bottom of the flask 
 is cherry-red for ten minutes. It is then allowed to cool 
 very gradually. 
 
 At this point a weighed quantity of ammonio-ferrous 
 sulphate, or ferrous sulphate of a known strength, is put 
 into the flask and then heated with rather dilute hydro- 
 chloric acid. The contents of the flask very soon dissolve ; 
 but it is well to keep shaking the solution while it is being 
 heated, to prevent loss of chlorine. It only remains now 
 to estimate the iron left unoxidised, in order to arrive 
 
430 ESTIMATION OF TITANIUM IN IRON. 
 
 at the quantity of manganese, which can be done with 
 potassium bichromate solution. It is feared that the ferrous 
 solution may get oxidised by exposure to the air ; a small 
 piece of marble put into the flask, which can also be fitted 
 with a cork and tube, will readily prevent that. 
 
 In four successive experiments the following results 
 were obtained : 
 
 No. Fe Equal to 
 
 Oxidised Manganese p. c. 
 
 1 0-2018 19-82 
 
 2 0-2103 20-65 
 
 3 0-2396 23-53 
 
 4 0-2435 23-88 
 
 No. 2 gave by ammonium acetate method 20-55 per 
 cent., which was done with great care. No. 4 is a repeti- 
 tion of No. 3. 
 
 It is evident, of course, that there is nothing original or 
 new in the above method ; but it contrasts very favourably 
 with the usual methods of separating the iron with sodium 
 or ammonium acetate, and precipitating the manganese 
 from the filtrate with bromine. It is not at all trouble- 
 some, does not take long, and has the advantage that the 
 only chemicals and apparatus required are those which 
 are necessary for the assay of iron ores. 
 
 Estimation of Titanium in Iron. The detection of 
 titanium in iron is easy, although its estimation is difficult. 
 The best results have been obtained by following Kiley's 
 plan.* This is essentially as follows : A weighed portion 
 of the iron borings are treated with fuming nitric acid in a 
 flask, a few drops of hydrochloric acid added from time to 
 time, the whole being well boiled. The contents of the 
 flask are then transferred to a porcelain dish, evaporated to 
 dryness, and heated strongly. On cooling, it will be found 
 that the iron oxide readily detaches itself from the dish, 
 and can be easily transferred to a beaker, the portions 
 left on the dish being dissolved in hydrochloric acid, and 
 poured on the contents of the beaker ; the dish may be 
 rinsed out, if necessary, with strong hydrochloric acid. 
 The contents of the beaker are boiled for from two to three 
 
 * Chemical News, viii. 226, 233. 
 
ESTIMATION OP TITANIUM IN IEON. 431 
 
 hours, until complete solution of the iron is effected ; and 
 as some quantity of hydrochloric acid is required for this, 
 the best plan is to allow a large portion of the excess of 
 acid to evaporate in the beaker, retaining only as much 
 as is requisite to keep the iron in solution. The silica 
 is filtered off in the usual way, after diluting with water 
 and adding a few drops of hydrochloric acid on the filter 
 to dissolve the basic salt formed by the water. By this 
 means the silica can be obtained nearly white after burning 
 off the graphite, and very little iron will be found with it 
 unless much phosphorus be present, as the silica invariably 
 contains more or less iron phosphate from the insoluble 
 iron phosphide, which cannot be completely dissolved out 
 by hydrochloric acid. Before estimating the titanium 
 the residue from the silica should be fused with potassium 
 bisulphate, dissolved in water, and added to the solution of 
 iron in which the titanium is to be estimated. The solu- 
 tion is reduced with sodium sulphite, and the excess of 
 sulphurous acid is driven off by boiling. The solution is 
 then near]y neutralised with ammonia, and ammonium or 
 sodium acetate added ; if there is only a small quantity of 
 phosphoric acid, there will always be sufficient peroxide 
 of iron to precipitate it, but if not, a few drops of nitric 
 acid must be added so that the precipitate produced is 
 distinctly red, and the solution boiled and filtered as 
 quickly as possible. The residue is fused with potassium 
 bisulphate, or, where nitric acid is used, this is driven off 
 with sulphuric acid. The result of the fusion with potas- 
 sium bisulphate is dissolved in cold water (when a little 
 iron phosphate, which remains insoluble, is separated), 
 boiled for some hours, and allowed to stand a night in a 
 warm place, when the titanic acid is filtered off and 
 washed with dilute sulphuric acid, dried, ignited, and 
 weighed. 
 
 The above process is not very satisfactory for the 
 quantitative estimation of titanic acid. The iron phos- 
 phate (insoluble in the potassium bisulphate) cannot be 
 washed without its passing through the filter ; and very 
 frequently, also, the small amount of iron keeps up the 
 
432 ESTIMATION OF TITANIUM IN IRON. 
 
 titanic acid, as iron even in small quantities has a very 
 great effect in preventing the precipitation of titanic acid ; 
 so that it is always advisable to add a little sodium 
 sulphite, which reduces the iron oxide and facilitates the 
 precipitation of the titanic acid. 
 
 Titanium may, however, be found more satisfactorily 
 and more readily during the process usually adopted to 
 estimate the amount of graphite in pig iron, provided a 
 large quantity of the pig be operated on. About 200 
 grains of the pig are to be dissolved in dilute hydrochloric 
 acid ; when the pig is nearly all dissolved, and the action of 
 the acid has ceased, more hydrochloric acid is added, and 
 the solution well boiled, so as to thoroughly extract all the 
 iron. The solution is then thrown on dried counterpoised 
 filters encircling each other, and the filter well washed to 
 remove all the iron. It is then treated with dilute potash, 
 and washed once ; then re-treated with it so as to entirely 
 remove the silica. The potash is thoroughly washed out, 
 and the filter treated with hydrochloric acid, thoroughly 
 washed, and dried at 250 F. until the weight is constant. 
 This gives the graphite, on burning which a residue of a 
 dirty light brown colour is left, which, fused with potas- 
 sium bisulphate, and subsequent treatment as above ex- 
 plained, is seen to be nearly pure titanic acid. 
 
 Mr. W. Bettell (' Chemical News,' Aug. 22, 1873) pro- 
 poses the following modification of Mr. D. Forbes' process 
 (' Select Methods in Chemical Analysis,' second edition, pp. 
 194, 195) for the estimation of titanic acid, which may not 
 be unacceptable to those engaged in the analysis of titanic 
 ores : 
 
 Fuse about 0-5 grm. of the finely powdered ore with 
 6 grms. of pure potassium bisulphate (which has been 
 recently fused and powdered) in a platinum crucible at a 
 gentle heat, carefully increased to redness, and continued 
 till the mass is in a state of tranquil fusion. 
 
 Eemove from the source of heat, allow to cool, digest 
 for some hours in 5 or 6 oz. of cold distilled water (not 
 more than 10 oz. is to be used, as it generally causes a 
 precipitation of some titanic acid) ; filter off from a little 
 
ESTIMATION OF HAKDNESS. 
 
 433 
 
 pure white silica, dilute to 45 or 50 oz., add sulphurous acid 
 till all the iron is reduced, then boil for six hours, replac- 
 ing the water as it evaporates. 
 
 The titanic acid is precipitated as a white powder, 
 which is now to be filtered off, washed by decantation, 
 a little sulphuric acid being added to the wash-water to 
 prevent it carrying titanic acid away in suspension. Dry, 
 ignite, allow to cool, moisten with solution of ammonium 
 carbonate, re-ignite, and weigh. The titanic acid is in- 
 variably obtained as a white powder, with a faint yellow 
 tinge, if the process has been properly carried out. 
 
 This method of fusing with potassium bisulphate 
 (' Select Methods in Chemical Analysis,' 2nd edition, p. 
 212) is preferable to all others for decomposing difficultly 
 soluble iron ores. 
 
 The Hardness of Iron and Steel. 
 
 Mr. T. Turner has given the following useful table 
 of the hardness of different varieties of iron and steel in 
 comparison with that of other bodies. The figures repre- 
 sent the weight in grammes necessary to produce a scratch 
 with a diamond on drawing its point over the smooth sur- 
 face of the metal. 
 
 Substances 
 
 Steatite 
 
 Lead (commercial) 
 
 Tin . 
 
 Eock salt 
 
 Zinc (pure annealled) 
 
 Copper (pure anne 
 
 Calcite . 
 
 Softest iron 
 
 Fluor-spar 
 
 Mild steel 
 
 Tyre steel 
 
 Good cast iron 
 
 Bar iron 
 
 Apatite . 
 
 Hard cast-iron scrap 
 
 Window -glass 
 
 Good razor-steel 
 
 Very hard white iron 
 
 
 
 
 
 Relative 
 
 
 
 
 
 hardness 
 
 
 
 ' 
 
 
 1 
 
 ) 
 
 
 
 
 1 
 
 
 
 
 
 2-5 
 
 
 
 
 
 4 
 
 ed) 
 
 
 
 
 6 
 
 aUed 
 
 
 t 
 
 
 8 
 
 
 
 
 
 12 
 
 
 
 
 
 15 
 
 
 
 * 
 
 L - 
 
 19 
 
 
 
 
 
 21 
 
 
 
 
 
 
 20^24 
 
 
 
 
 - - 
 
 2124 
 
 
 
 
 
 24 
 
 
 
 
 
 34 
 
 rap 
 
 
 
 
 36 
 
 
 
 
 
 60 
 
 \ 
 
 
 
 
 60 
 
 ron 
 
 
 
 
 72 
 
 F F 
 
434 
 
 CHAPTER X. 
 
 THE ASSAY OF COFFEE. 
 
 IN the assay of copper by the dry way, all minerals and 
 substances containing that metal may be divided into 
 three classes. 
 
 CLASS I. Comprises Sulphuretted Ores or Products, with 
 or without Selenium, Antimony, or Arsenic. 
 
 Copper glance, Cu 2 S, containing 79'7 p. c. of copper 
 
 Chalcopyrite, Cu 2 S, Fe 2 S 3 , 34-4 
 
 Erubesoite, 3Cu 2 S, Fe 2 S 3 , 55'7 
 
 Bournonite, 3Cu 2 S,SbS 3 + 2(3PbS,SbS 3 ) 12-7 
 
 Fahlerz, 4(Cu 2 S,FeS 5 ZnS,AgS,HgS).(SbS3,AsS 3 ,Bi 2 S 3 ) 
 
 3048 
 
 Covelline CuS, 66'7 
 
 Wolfsbergite, Cu 2 S,SbS 3 , 24'9 
 
 Domeykite, Cu 6 As, 71'6 
 
 Copper regulus, Copper speiss, &c. 
 
 CLASS II. Oxidised Ores and Products. 
 
 Red copper, CojO, containing 88'7 per cent, of copper 
 
 Malachite, 2CuO, + H 2 0, 57'3 
 
 Azurite, 2CuO,C0 2 + CuO,H 2 0, 55-1 
 
 Cyanosite, CuO,S0 3 +5H 2 Q, 25'3 
 
 Phosphate of copper, 30 56 
 
 Arseniate of copper, , , 25 50 
 Chromate, Vanadate, and Silicate of Copper ; Slags, &c. 
 
 CLASS III. Copper and its Alloys. 
 
 The different methods of assaying copper are more 
 numerous than those for any other metal. They are in some 
 cases similar to each other, and in others based upon very 
 different principles. 
 
CLASSIFICATION OF THE COPPER ASSAYS. 435 
 
 These methods may be divided into : 
 
 A. ASSAY IN THE DRY WAY. 
 
 a. For Rich Ores and Products of Class I. 
 
 I. English Copper Assay. 
 
 b. For Ores and Products of Class II. 
 
 1. Lake Superior Fire Assay, 
 
 B. ASSAYS IN THE WET WAY. 
 
 a. Colorimetric Copper Assay. 
 6. Volumetric Copper Assay. 
 c. Electrolytic Copper Assay. 
 
 A. ASSAYS IN THE DRY WAY. 
 a. For Rich Ores and Products of Class I. 
 
 I. ENGLISH COPPER ASSAY. 
 
 M. L. Moissenet has given, in the 'Annales des Mines,'* 
 a very complete description of the Cornish method of 
 assaying copper by the dry way. The following is from a 
 translation by Mr. W. W. Procter. 
 
 Each of the large Swansea copper-works keeps an 
 assayer at Cornwall, whose duty it is to estimate the 
 richness in copper of all the lots of minerals in the county 
 sold every Thursday at the Ticketing, and of all the 
 samples of foreign minerals and copper products which 
 may be useful to the smelter. 
 
 The copper being obtained in the state of a prill or 
 metallic button, the impurities (generally tin, antimony, 
 &c.) are thus made evident, and the hammer soon proves 
 the quality of the metal which we ought to expect to 
 obtain by metallurgic treatment. As for the accuracy of 
 the method, as far as regards the whole of the metal 
 obtained, we shall revert to this later on. We would, 
 however, observe that, within certain limits, the method 
 would not be less practical on account of being inexact ; 
 
 * Vol. xiii. p. 183. 
 
 p p 2 
 
436 ENGLISH COPPER ASSAY. 
 
 for we must not forget that it has chiefly for its object 
 to teach the smelter the value of the mineral, even more 
 than its true richness. 
 
 For example, if we get too low an assay from a sample 
 of 2 or 3 per cent., we should only from this assent to the 
 opinion of the metallurgist, whose interest it is not to work 
 upon very poor minerals. The same remark will apply to 
 the case of minerals very antimonial, &c. Besides, in the 
 description of the method we shall discover the principal 
 phases of the Welsh process ; so that it is more just to con 
 sider the Cornish assay as a metallurgy on a small scale than 
 as a scientific laboratory method. From thence result also 
 the necessity of long practice and the almost uselessness of 
 theoretical knowledge-for those who purpose employing 
 this method alone. 
 
 Sir Henry de la Beche (' Eeport on the Geology of 
 Cornwall,' &c., p. 595), in giving a sketch of the method, 
 declares it to be rather rough and uncertain, and fails not 
 to add at the conclusion a translation of a passage relative 
 to the assay of copper pyrites from M. Berthier's treatise 
 on assays by the dry way. 
 
 These drawbacks upon the scientific value of the Cornish 
 method cannot injure the power of facts ; they constitute 
 but another reason which we may have for giving an ac- 
 count of the manner in which the first basis of the valua- 
 tion of the greater part of the copper minerals has been 
 fixed since so long a period. 
 
 Division Adopted. The rather complex operations 
 through which we have to pass will be better apprehended 
 by explaining in succession 
 
 1. The order of the operations, the nature and influence 
 of the fluxes employed, the kind of products obtained 
 (reactions). 
 
 2. The manipulations to which each operation gives 
 rise, the furnaces and apparatus used, the characters of the 
 principal products during the chief phases and at the end 
 of each (manipulations). 
 
 We shall add to these 
 
 3. Some information upon the influence of the princi- 
 
ENGLISH COFFEE ASSAY. 437 
 
 pal foreign metals (tin, antimony, zinc, lead), and upon the 
 treatment of some special coppery matters. 
 
 4. Summary considerations on the result of the English 
 method compared with those of the analysis by the wet 
 way. 
 
 SECTION I. EEACTIONS. 
 
 At the very outset we distinguish two kinds of assays. 
 
 1. The roasted sample. 
 
 2. The raw sample. 
 
 The first only applies to cupreous pyrites or to samples 
 essentially formed of it that is to say, which contain 
 sulphur in excess ; the process begins by a roasting. 
 
 In the raw assay we dispense with the roasting ; we 
 have recourse to the addition of reagents, either oxidising 
 or sulphurising, according to the minerals ; we endeavour 
 to place them by these mixtures in the condition of a pro- 
 perly roasted pyritic mineral. 
 
 From this point, at least in general, the operations 
 become identical. They consist in 
 
 1. Fusion for regulus (regulus). 
 
 2. Calcining the regulus (calcining). 
 
 3. Fusing for coarse copper (coarse copper). 
 
 4. One or two fusions with fluxes (washings). 
 
 5. Trial by striking with a hammer, last refining (test- 
 ing, refining). 
 
 6. Treatment of slags for prill. 
 
 All the slags except those of the fusion for regulus have 
 been preserved. The fusion No. 6 gives a small sup- 
 plementary button of copper, which again undergoes, if 
 necessary, one or two washings. 
 
 As we have said, the roasting is used only for pyrites. 
 We shall return later on to the duration and the circum- 
 stances of this operation. Its evident aim is to drive off 
 the excess of sulphur, so as to cause the whole of the 
 copper, with a part only of the iron which abounds in the 
 pyrites, to pass into the state of sulphide at the time of 
 the fusion for regulus. 
 
438 ENGLISH COFFEE ASSAY. 
 
 I. Eegulus. 
 
 1. Pyrites. The fusion for regulus of a properly 
 roasted pyrites is made by mixing with it equal volumes 
 of the three fluxes borax, fluor-spar in powder, lime 
 slaked in powder of each one ladle, and covering the 
 mixture with a layer of moist common salt. The matters 
 composing the gangue of the roasted mineral consist prin- 
 cipally of quartz, silica, and in general of more alumina 
 and magnesia than lime ; oxide of iron, resulting from the 
 roasting of the pyrites, is also present. 
 
 The borax only serves to give fusibility, the fluor-spar 
 contributes to the same end by forming a fluosilicate. 
 Otherwise it does not play an important part in the de- 
 composition that is to say, there is probably no produc- 
 tion of fluoride of silicon and calcium ; for this last base is 
 added here in considerable proportion, so as to form im- 
 mediately a silicate which may combine with the fluoride 
 of calcium. 
 
 The ferric oxide being so reduced as to pass into the 
 slag, and the different metallic oxides to pass into the re- 
 gulus, yield oxygen, which reacts on the remaining sulphur. 
 The disengagement of sulphurous acid which results from 
 this, joined to the water contained in the fluxes, justifies 
 to a certain extent the use of a layer of common salt, 
 designed to prevent the boiling over. Besides this, the 
 common salt, being without action on the metallic sul- 
 phides, does not here produce those important effects 
 which it exerts in the later fusions. 
 
 If the pyrites appear insufficiently roasted, we must 
 add a little nitre, the oxidising action of which again gives 
 off sulphur ; the opposite case, that of a roasting too 
 much prolonged, is rare ; we remedy it by the addition of 
 sulphur and tartar. 
 
 2. Very poor Pyrites. In a very poor pyrites that 
 of Bear Haven, in Ireland, for example the proportion of 
 sulphur does not require us to have recourse to the roast- 
 ing ; we employ the three fluxes and one ladle of nitre. 
 
ENGLISH COPPER ASSAY. 439 
 
 3. Variegated Copper Ore. Peacock ore contains less 
 sulphur in proportion to the copper than pyrites ; we also 
 fuse with a little nitre. 
 
 4. Sulphide of Copper. The sulphur is here insuffi- 
 cient. We add together sulphur -J to 1 ladle, according 
 to the valuation ; tartar J to ^ ladle that is to say, half 
 the volume of the sulphur. The tartar is a powerful re- 
 ducing agent, and is supposed in small quantities to favour 
 the action of the sulphur by preventing its disengage- 
 ment as sulphurous acid by the oxidising matters in the 
 mineral ; but if used in excess, it acts as a desulphuriser, 
 as well by its carbon as by its alkali. 
 
 5. Carbonated Minerals. The addition of sulphur and 
 carbon is evidently still more necessary here. 
 
 6. Native Mixture : f sulphide of copper, ^ pyrites. 
 We add, in this case, nitre for the pyrites, and sulphur and 
 tartar for the sulphide of copper ; although these reagents 
 appear sure to neutralise each other, it is possible that their 
 simultaneous employment may be logical. The nitre pro- 
 bably decomposes the pyrites, which would without it fuse 
 and give a very ferrous regulus, whilst the free sulphur 
 would be of little use, on account of the sulphide of copper. 
 Be this as it may, this is the plan adopted. 
 
 During the progress of the fusion for regulus we have 
 still to introduce other matters, some incidentally, and 
 others in all cases. 
 
 If a blue flame persists in escaping from the crucible, 
 an index of the formation of sulphurous acid, we project 
 into it sulphur 1 ladle, tartar ^ a ladle. When the fusion 
 appears almost finished, in order to render the bath more 
 liquid, and to facilitate the collection of the button, we 
 throw in a little dried salt, and a flux composed before- 
 hand of lime, a little fluor-spar, and a very little borax 
 that is to say, of the elements in different proportions of 
 the mixture introduced originally. 
 
 The regulus obtained is composed principally of copper, 
 iron, and sulphur. We shall return to the aspect and the 
 richness which it ought to have according to the minerals 
 treated. 
 
440 ENGLISH COPPER ASSAY. 
 
 II. Calcining. 
 
 The calcination of the regulus is one of the most im- 
 portant operations ; it ought to be quite complete. 
 
 III. Coarse Copper. 
 
 To the calcined regulus is added nitre J ladle, borax 
 
 1 ladle, charcoal -J ladle, dry salt 1 ladle (these quantities 
 remain the same, whatever mineral may be assayed) ; tartar 
 
 2 ladles for a regulus of medium richness. Covering of 
 moist salt, 2 ladles. 
 
 The nitre is designed to burn the sulphur which may 
 have escaped the calcining, and to insure the passage of 
 the easily oxidisable metals, especially of iron, into the 
 slag in the state of oxides. It is, besides, in too small pro- 
 portion to act upon the copper, especially in presence of 
 reducers whose effect is certainly later than the deflagra- 
 tion of the nitre. 
 
 The borax plays simply the part of a flux. 
 
 The dry salt has for its object to give fluidity to the 
 slag. Unfortunately, if the addition of the salt attains this 
 object, it also determines from this operation a sensible 
 loss of copper by carrying it away with the saline vapours. 
 We shall insist upon this point in describing the washing. 
 
 The charcoal and the tartar are especially the important 
 reagents in the fusion. The tartar, at the same time that 
 it is one of the most energetic reducers, is also a flux and 
 a desulphuriser. Its use is, then, perfectly justified here, 
 only the proportion of tartar added ought to be regulated 
 according to the quantity of copper, which the weight and 
 aspect of the regulus permit the experienced assayer to 
 estimate sufficiently closely ; an excess of tartar would re- 
 duce the foreign metals, and produce in consequence a 
 very impure coarse copper. 
 
 When the fusion appears complete, we throw in some 
 white flux,* which gives fluidity to the slag, and deter- 
 
 * This white flux is prepared in the laboratory by mixing in a mortar, 
 tartar 3 volumes, nitre 2 volumes, salt a little, then determining the combus- 
 tion by the introduction of a red-hot iron rod, which is turned round until the 
 matter ceases to deflagrate. 
 
ENGLISH COPPER ASSAY. 441 
 
 mines by its partial decomposition, from which a disen- 
 gagement of carbonic oxide results, a stirring up of the 
 materials. These two effects facilitate the collection of 
 the metallic button. The potassium carbonate begins also, 
 without doubt, from this operation to refine the metal a 
 little by attacking the iron, zinc, and tin already reduced. 
 M. Berthier (' Essai par la Yoie Seche,' vol. i. p. 393) 
 points out this reaction : ' A part of the carbonic acid 
 which it contains being decomposed and changed into car- 
 bonic oxide, a compound is formed consisting of alkali, 
 carbonic acid, and metallic oxide, &c. 
 
 Lead, copper, and antimony are not attacked. 
 
 IV. Washings. 
 
 In the operation of washing we put into the crucible, 
 at the same time as the coarse copper, the following 
 fluxes : White flux, 1 ladle ; dry salt, 2 ladles. 
 
 It is evident that the white flux is here employed as 
 an oxidiser of the foreign metals, and with a view to the 
 application of the above-mentioned reaction. 
 
 As for the salt, it is both useful and injurious. If it were 
 only used with the view of augmenting the fluid mass so as 
 to preserve the metal from contact of air, &c., it would be 
 advantageously replaced by an excess of white flux ; but it 
 can form volatile chlorides with the arsenic and antimony 
 which the copper has retained in the form of arsenide 
 and antimonide. Common salt is, then, to be regarded as 
 one of the principal agents of purification put in operation 
 by the Cornish method. On the other hand, the loss of 
 copper which arises from the carrying off of this metal by 
 the vapours of common salt cannot be doubted. M. Ber- 
 thier has found that by heating equal weights of copper 
 and salt until the complete volatilisation of the latter, 3 
 per cent, of the metal is carried off. 
 
 In the event of the coarse copper appearing too im- 
 pure, we take care to add a little nitre. According to the 
 appearance of the button we recommend the washing or not. 
 
442 ENGLISH COPPER ASSAY. 
 
 Y. Testing, Refining. 
 
 The button of metal is flattened on an anvil. We 
 thus recognise tin by the hardness, and antimony by the 
 brittleness, of the alloy. The button is then put alone in 
 the crucible. When it presents a proper appearance 
 that is, when the edges assume a bright colour, the centre, 
 which the assay er calls the eye, being dark ^we hasten to 
 put into the crucible the fluxes, which are the same as 
 for washing, only taken in rather smaller quantity. 
 
 In general, when we have operated well, the button 
 obtained is of a fine colour, and is regarded as pure ; if we 
 have passed the eye, it is covered with a layer of red 
 oxide ; if, on the contrary, we have put in the fluxes too 
 soon, the button is dull. 
 
 It is easy to give an account of the reactions which 
 take place during the refining, and which differ a little 
 from those of the washing. 
 
 In heating the button alone in the air in the crucible, 
 it is intended to submit it to an oxidation which ought to 
 act sufficiently on all the foreign metals more oxidisable 
 than copper without acting too much on the latter. The 
 proper point is indicated by the appearance of the eye ; 
 the projection of the fluxes puts an end to the atmos- 
 pheric oxidation, and causes the scorification of the oxides 
 which expel part of the carbonic acid of the potassium 
 carbonate, for which they substitute themselves, and give 
 rise to triple compounds of metallic oxides, alkali, and 
 carbonic acid. 
 
 The lead, tin, iron, and zinc oxides behave thus. 
 When we have passed the eye, there has been a con- 
 siderable formation of oxide, which leaves the button 
 reddened, as I have indicated. At the same time the slag 
 is strongly coloured red or green. If, on the contrary,, 
 the fluxes have been thrown in too soon, the oxidation 
 has been insufficient, and then the refiner just falls back 
 upon the preceding operation of washing an operation 
 less efficacious, and even without result, in the case of lead 
 and antimony. 
 
ENGLISH COPPER ASSAY. 443 
 
 As for the physical phenomenon of the eye, perhaps it 
 corresponds to the very short instant when the oxides, less 
 dense than the copper, are concentrated at the top of the 
 button, .and there make a dark spot before attaining a 
 temperature sufficiently elevated to acquire the brightness 
 of the metal itself. 
 
 We may add that the minerals of Cornwall, generally 
 more impure than foreign minerals, require a notably 
 longer time for the appearance of the eye. 
 
 Extra Accidental Washing. More often the refining 
 gives a definite product, put aside to be weighed with the 
 prill extracted from the slag ; let the button be clear, burnt, 
 or dull. Even if the metal appeared too impure we would 
 not recommence the refining, but would have recourse to 
 an extra washing by putting at once into the usual crucible, 
 besides the button and the usual fluxes, the slag from the 
 refining. 
 
 VI. Slags for Prill. 
 
 All the slag from the fusion for coarse copper in- 
 clusively having been preserved, we fuse them altogether 
 with : 
 
 Tartar . . .1 ladle "1 Simple reducing 
 Charcoal . . . traces / mixture. 
 
 We obtain a small globule, variable with the circum- 
 stances of the different operations which have allowed 
 more or less copper to pass into the slag. If the prill is 
 not very small, and its appearance indicates a metal not 
 sufficiently pure, we submit it to one or two washings as 
 above. 
 
 SECTION II. MANIPULATIONS. 
 
 The sample, which has been taken with the utmost 
 care, arrives at the laboratory rather coarsely powdered, 
 still wet, and wrapped in strong packing-paper ; the paper 
 is opened and placed near a furnace on the cast-iron plate 
 which covers it ; the drying is rapidly done there. 
 
 The first requisite is to discover the kind or kinds of 
 minerals, so as to employ the warm or raw sample. 
 
444 ENGLISH COPPEE ASSAY. 
 
 For this purpose we throw one or two large pinches of 
 the mineral into a flat-bottomed copper dish, and we wash 
 it very easily by putting in water several times and giving 
 a rotatory motion to the matters at the same time that 
 we incline the dish so as to cause the muddy parts to run 
 from the gangue. The small metallic fragments remain 
 distinctly visible, and we can often discern by simple in- 
 spection the presence of foreign metals. 
 
 We weigh 400 grs. of the dried mineral a quantity 
 upon which the assay is made. 
 
 The crucibles used in Cornwall are of three sizes : 
 
 1. Large. 
 
 2. Large second. 
 
 3. Small second. 
 
 The small seconds have externally the internal dimen- 
 sions of the large, into which they fit as into a nest ; the 
 first and third are sold the one in the other, and called 
 nested. They are the most used. 
 
 The large serve for the roasting and the fusion for 
 regulus: the small seconds for calcining the regulus and 
 all the fusions which follow. 
 
 The large seconds are only employed in place of the 
 former when we have to treat a very large regulus. 
 
 The crucibles are of a kind rather wrinkled, and as if 
 fused superficially ; they present the appearance of coarse 
 stoneware pottery. Their form, moderately wide, permits 
 us to make use of them successively for the roasting and 
 the fusion for regulus, and gives them sufficiently great 
 stability in the fire of a wind-furnace. They are, besides, 
 very resisting. They are made at Truro and Eedruth. 
 
 The wind-furnace has for its principal dimensions 
 
 Inches 
 
 Length from front to flue . . . 10 
 
 Breadth 
 
 Depth to the bars . 
 
 Opening of the flue 
 
 8 
 
 14 
 
 8 
 
 2 
 
 A sufficiently large space is reserved underneath the 
 fire, where the ashes accumulate without inconvenience, 
 but opening only by a contracted framework so as not to 
 allow too free an access of cold air. 
 
ENGLISH COFFEE ASSAY. 445 
 
 The furnace serves either for roas tings or for fusions ; 
 in the latter case we cover it with two mounted bricks, very 
 easy to manage, and allowing us to only half-open it when 
 we wish to inspect the contents of the crucibles. We can 
 conduct ten roastings at once ; the crucibles are marked by 
 a brush with colcothar mixed with water. The furnace 
 having been recharged with coke, we put the crucibles on 
 the top, and after a few minutes, the substances beginning 
 to get warm, we stir them by means of iron rods. Each 
 crucible receives a rod, which we leave standing there 
 (leaning against the chimney) during the whole period of 
 the roasting, so as to avoid the loss which would take 
 place if we withdrew the rod. From time to time we 
 renew the surfaces by lightly taking hold of the rod with 
 the left hand by the upper end, whilst the right forefinger 
 and thumb make it turn at once upon itself and round the 
 crucible. 
 
 The duration of the roasting varies essentially with the 
 nature and the richness of the mineral ; it is never less 
 than six or seven minutes, and may reach half an hour. 
 When from the sandy appearance of the matters we con- 
 sider the operation finished, we withdraw the crucible, 
 raise the iron rod with care, and expose the crucible to the 
 air, allowing its contents to cool slowly. The roasting has 
 succeeded when the surface has the brown-red colour of 
 iron oxide and the bottom only is black. In this case we 
 proceed to the fusion for regulus by simply adding the 
 three fluxes (borax, fluor-spar, and Jime) ; if the bottom 
 of the crucible appear too black, we ought to complete 
 the oxidising action by the addition of a little nitre. 
 
 Fusion for Regulus. 
 
 The different substances above indicated are taken 
 from the box with a slightly concave ladle of If diameter, 
 then mixed in the crucible with a stirring-knife. We 
 ought to allow the heat of the wind-furnace to fall and 
 then to recharge, so as to have a gentle fire at the com- 
 mencement of the fusion for regulus. The crucibles 
 
446 ENGLISH COPPER ASSAY. 
 
 placed upon the coke, and supported against the walls of 
 the furnace, which we then close with the two bricks. 
 After about a quarter of an hour, we open the front brick, 
 so as to observe the progress of the operation ; it is at this 
 stage that we throw the sulphur and tartar into those 
 crucibles from which a blue flame is disengaged. Some 
 minutes later that is to say, nearly seventeen minutes 
 from the commencement we add the salt and the flux 
 destined to collect the regulus ; then (twenty minutes from 
 the beginning) we run into a metal mould, not greased. 
 
 We make, in general, several fusions at once four, 
 for example ; we have in consequence two moulds into 
 which we pour the contents of the crucibles in an adopted 
 order, so as to avoid all confusion. The matters, very 
 rapidly solidified, are detached simply by a blow, and fall 
 in order on a metal plate fixed in front of the laboratory 
 window. We immediately seize them with the copper 
 tongs, put them into a basin of the same metal, and im- 
 merse them for a moment in cold water, where it is 
 important not to leave them too long. This immersion 
 allows us then to separate very easily the slag from the 
 button of regulus, itself very brittle. For this purpose the 
 mass is put on the metal plate, and by means of a hammer 
 we strike with care all round the slag, which breaks off 
 pretty cleanly. We hasten to detach from the surface of 
 the regulus the slag which may remain adherent, using a 
 small hand-chisel, without the hammer. The slags are 
 broken, and if we find any prills of regulus they are added 
 to the principal button. Sometimes in these breakings, 
 and especially in those analogous for the last fusions, we 
 surround the substances by an iron ring, placed on the 
 metal plate, so as to avoid loss of splinters. In a general 
 way, the slags of the fusion for regulus are rejected. We 
 shall see further on how it may become necessary to flux 
 them again when the mineral contains blende. 
 
 The aspect of the regulus is characteristic, and it is 
 easy to arrive at a pretty close estimation of its richness, 
 and consequently of the degree of success of the operation, 
 by simple inspection of the regulus. 
 
ENGLISH COPPER ASSAY. 
 
 447 
 
 No. 1 . A very poor regulus (coarse) ; that is to say, too 
 much charged with iron, is bronzed and dull; the opera- 
 tion following would not be able to carry off the excess 
 of iron, at least without a corresponding loss of copper. A 
 like regulus evidently results from an imperfect warming, 
 or from an excess of sulphur, or from an insufficiency of 
 nitre, as the case may be. 
 
 It contains less than 40 per cent, of copper. There is 
 nothing for it but to reject it. 
 
 No. 2. A regulus of good appearance is in general 
 bronzed but rather shining ; it appears finer. Its richness 
 varies from 40 to 60 per cent. 
 
 No. 3. From oxides, carbonates, and from some mine- 
 rals charged with impurities (Sn, Sb) we desire to obtain 
 a fine bluish button of a greater richness 65 to 75 per 
 cent. We perceive, indeed, that from oxides and carbon- 
 ates, to which we have only to add sulphur, and which 
 also by their nature do not, like pyrites, contain combined 
 iron, it is easy to obtain a richer regulus without fearing 
 any loss of copper. As for the stanniferous and antimonial 
 minerals, we shall return to them further on. 
 
 No. 4. In every case a regulus, the richness of which 
 rises to 80 per cent., and of a very shining grey-blue 
 appearance, ought to be rejected, its richness indicat- 
 ing the loss of a certain quantity of copper left in the 
 slag. 
 
 Here is, in the preceding order, the result of the ana- 
 lyses of four buttons whose description agrees with that 
 which we have just given, excepting, perhaps, No. 2, 
 whose fracture is rather reddish : 
 
 No. 
 
 
 Copper 
 
 Iron 
 
 Balance ; 
 sulphur 
 and traces 
 of foreign 
 metals 
 
 1 
 
 2 
 3 
 4 
 
 Coarse, to be rejected . 
 Good in general (rather too fine) . 
 Good for a carbonate, &c. . 
 Too fine, to be rejected 
 
 36-00 
 60-00 
 65-60 
 80-16 
 
 32-90 
 14-70 
 10-50 
 2-10 
 
 31-10 
 25-30 
 13-90 
 17-74 
 
 If we compare these products with those obtained in 
 
448 
 
 ENGLISH COPPER ASSAY. 
 
 tlie metallurgy of copper by the Welsh method, we find 
 (Le Play, ' Annales des Mines ') : 
 
 Matts of the operations 
 H. V. IV. VIII. 
 
 h 
 
 Si 
 
 a 
 
 a 
 
 d 
 
 2 
 
 (-H 
 
 Different 
 
 metals 
 
 1 
 a 
 "9 
 m 
 
 co 
 
 3 
 
 
 
 ( Coarse matt (fusion of poor mine- j 
 
 
 
 
 
 
 II. \ rals, raw or calcined) 
 
 34-6 
 
 34-1 
 
 1-5 
 
 29-8 
 
 31-3 
 
 UCu 2 S + Fe 2 S 3 + 4(Fedif.met.)S j 
 
 
 
 
 
 
 I/ Blue matt (fusion of the cal- \ 
 
 
 
 
 
 
 cined coarse matt with I 
 minerals of mean richness) [ 
 
 57-2 
 
 18-5 
 
 1-0 
 
 23-3 
 
 24-3 
 
 1 0-8Cu + 3Cu 2 S + 2(Fe. d.m.)S J 
 
 
 
 
 
 
 1 Beddish variety, matte mince } 
 \ l-3Cu + 3Cu 2 S + 2(Fe. d.m.) S ) 
 
 61-6 
 
 15-8 
 
 0-6 
 
 22-0 
 
 22-6 
 
 /White matt (fusion of the calcined 
 
 
 
 
 
 
 I course matt with rich minerals, 
 
 
 
 
 
 
 carbonates, and oxides) 
 
 
 
 
 
 
 IV J Metal very pure type 
 very blue variety . 
 
 77.4 
 64-8 
 
 0-7 
 9-0 
 
 0-9 
 3-6 
 
 21-0 
 22-6 
 
 21-9 
 26-2 
 
 mean . 
 
 73-2 
 
 6-3 
 
 
 
 20-5 
 
 
 
 8Cu 2 S + FeS 
 
 
 
 
 
 
 (Matt (roasting of extra white] 
 
 
 
 
 
 
 VIII A matt VII.) 
 
 81-1 
 
 0-2 
 
 
 
 18-5 
 
 
 
 ( O2Cu + Cu 2 S J 
 
 
 
 
 
 
 These numbers show the evident analogy, the identity 
 almost, of the products of the laboratory and those of the 
 works ; we may sum up by saying that the regulus ought 
 to be richer than coarse metal, and in the case of ordinary 
 minerals to approach if not to attain (as in the case of 
 sample No. 2) to the composition of blue metal. 
 
 For carbonated and oxidised minerals we arrive 
 directly at the very bluish variety of white metal. 
 
 Finally, in no case must we have a button as rich as 
 regulus matt. 
 
 Calcining the Matt. 
 
 The matt is pounded fine in a bronze mortar ; we avoid 
 loss of fragments by means of a perforated cover and a 
 cloth which surrounds the pestle. To facilitate the pulver- 
 isation, and avoid the sulphide greasing, we add in the 
 mortar a small piece of coke. The pounded matt is care- 
 fully turned upon a sheet of paper, the mortar wiped out 
 with a hare's foot, and the substance put into a small 
 second or large-second crucible. The calcining is con- 
 
ENGLISH COPPER ASSAY. 449 
 
 ducted as the roasting of a mineral ; it generally lasts 
 longer, for the expulsion of the sulphur is to be as com- 
 plete as possible. It is necessary to regulate the fire with 
 the greatest care, so as to avoid all agglomeration, and to 
 stir almost continually. When the matter adheres to the 
 rod, we withdraw the crucible for a moment ; this incon- 
 venience is chiefly produced if we have not detached 
 the slags sufficiently from the matte ; the calcining is then 
 much longer, the flames remain blue a long time, and the 
 fumes which are disengaged have an odour which is -not 
 purely that of sulphurous acid. When the fumes and the 
 odour cease, and the matter has taken a sandy appearance, 
 we raise the heat ; then withdraw the crucible, and allow 
 it to cool slowly in the air as when roasting. 
 
 The mean duration of calcining is half an hour. 
 
 Coarse Copper. 
 
 The fluxes above indicated are taken from a box No. 1, 
 except the dry salt called the refining flux, which forms 
 part of a second box. The ladle for this box No. 2 is a 
 little larger than for the first : it has a diameter of 1^ 
 inch. At the beginning of the operation the furnace is 
 well filled and lighted ; the same fire ought to suffice for 
 all the following fusions, which it is very important to 
 conduct with great rapidity. After a moment, and if there 
 is any frothing, we throw in some dry salt, which calms 
 the ebullition. At the end of ten minutes, the fusion ap- 
 pearing complete, we throw in a pinch of white flux. A 
 little after we withdraw successively each of the crucibles, 
 pouring them carefully and by a single turn into each of 
 the principal cavities of the metal mould. These moulds 
 ought, this time only, to be greased with a cloth impreg- 
 nated with suet. The crucibles are immediately put back 
 again into the fire. 
 
 We detach the slag as previously, seize' each one 
 successively with the copper tongs, and plunge it for an 
 instant into a basin full of water. The rest is effected as 
 for the regulus, only the slags are preserved on the metal 
 
 G G 
 
450 ENGLISH COPPER ASSAY. 
 
 plate, and in the order in which we have detached them. 
 The button of copper obtained appears more or less black ; 
 T have already indicated the influence of the tartar in 
 excess. 
 
 Washings. 
 
 We place the button and the fluxes in a large copper 
 shovel, lengthened and narrowed at the end, called a scoop, 
 and we pour them into the crucible, which is already at a 
 red heat. As the fusion is made in five or six minutes, it 
 would be inconvenient to prolong it on account of the loss 
 occasioned by the carrying off of copper with the vapours 
 of common salt. The pouring is made with care first into 
 one of the large cavities, then as soon as the metal has 
 fallen there we finish by casting the slag into one of the 
 small lateral cavities. This last slag, probably rich in 
 copper, is less fluid, and would adhere to the button, 
 which would be difficult to cleanse. The two buttons 
 being detached from the mould, we immerse the small one 
 first, then finish as in the preceding operation. 
 
 Testing and Refining. 
 
 The crucible has again been put back into the furnace, 
 after the pouring ; the button tried by the hammer is put 
 into the crucible by means of the tongs. At the end of 
 about three or four minutes it attains the colour of the 
 vessel, the eye manifests itself, and we rapidly throw in 
 the fluxes put into the scoop beforehand. 
 
 The pouring is done as for the washing, with the small 
 button of slag kept apart. 
 
 In general we get a button regarded as pure, clean 
 copper ; if not, as I have said, we proceed to an extra 
 washing by adding exceptionally in the scoop the last slag 
 obtained. 
 
 Prill. 
 
 The crucible this time has been left out of the furnace ; 
 put into it all the slags, collected for this purpose from 
 the metal plate into the scoop, and upon which we have 
 
ENGLISH COPPER ASSAY. 451 
 
 put reducing reagents. The fusion lasts a quarter of an 
 hour ; pour all at once into the large cavity : before the 
 cooling, by means of a transverse blow, get rid of the 
 upper beds which are still liquid, and composed principally 
 of common salt. Then operate as above. Collect the 
 prill, which again undergoes, if necessary, a washing. 
 
 SECT. III. SOME MINERALS AND SUBSTANCES OF A SPECIAL 
 NATURE INFLUENCE OF FOREIGN METALS. 
 
 STANNIFEROUS MINERALS. Most often we only perceive 
 the presence of tin in a copper mineral when testing with 
 the hammer, which reveals the nature of the bronze ; 
 when we proceed to the refining of such a stanniferous 
 button, it is impossible to obtain the characteristic eye ; 
 that is to say, the surface of the metal becomes quite clear, 
 and we scarcely open the furnace when it again becomes 
 obscure. We free it from tin by two or three extra wash- 
 ings. If we suspect tin from the known produce of the 
 mineral, or the inspection of the sample in the basin, we 
 endeavour to obtain a fine regulus, which is accomplished 
 in the case of a warm sample by prolonging the calcining, 
 and for the raw sample by putting in more nitre or less 
 sulphur. It is clear that tin can only enter the regulus by 
 virtue of the excess of sulphur necessary to the formation 
 of the coppery matte, and that by restraining this excess of 
 sulphur we diminish the chance of tin entering the button. 
 The fine regulus ought to contain 70 to 75 per cent, of 
 copper, as for the carbonated copper minerals. 
 
 ANTIMONIAL MINERALS. Antimony is also detected in the 
 testing, the metal being rendered very brittle, We then 
 add one or two grammes (15 to 30 grains) of lead in the 
 refining operation. There forms an alloy of lead and an- 
 timony heavier than copper, which is poured into the small 
 cavity of the mould. When we suspect antimony we act 
 as for tin that is to say, we produce a fine regulus, a 
 most careful roasting expelling the antimony ; then we 
 have to make two washings, and in the second to add the 
 metallic lead. 
 
 G G 2 
 
452 ENGLISH COPPER ASSAY. 
 
 We cause, then, three influences to act with a view of 
 expelling the antimony : 
 
 1. Slow oxidation at a low temperature, disengaging 
 antimony. 
 
 2. Eepeated chloridations, whence a formation of vola- 
 tile chlorides. 
 
 3. Affinity of the lead and mechanical separation of 
 the alloy. 
 
 ZINCIFEROUS MINERALS. One of the metals which is 
 most troublesome is zinc. We recognise it by the appear- 
 ance of the regulus and by its colour, which is that of 
 blende. Perhaps once out of ten the regulus collects suffi- 
 ciently to be able to detach it ; in this case we pound it, 
 add to it the slags, and borax 1 ladle, nitre ^ ladle. We 
 fuse anew, and obtain a good regulus, for the nitre has 
 caused the zinc to pass into the slag in the state of oxide. 
 
 Most often the zinciferous regulus does not collect, and 
 there is nothing for it but to begin anew by making a 
 very prolonged roasting of at least half an hour. 
 
 PLUMBIFEROUS MINERALS. Lead is not injurious, for it 
 does not alloy with copper. The warming is also pro- 
 longed. Lead passes into the regulus, which facilitates the 
 collection of the matter. In the last operation the lead 
 easily passes into the slag ; it also, in case of need, carries 
 off antimony. Thus the copper obtained from lead minerals 
 is most malleable. 
 
 Special Cupriferous Products. 
 
 EEGULUS OF CHILI. These are treated as those which 
 we obtain by the fusion for regulus. Their richness, 
 which rises to nearly 60 per cent., requires us to add 
 much tartar in the fusion for coarse copper. 
 
 SLAGS OF COPPER. To obtain regulus w^e add to the slag 
 sulphur, tartar, and nitre, this last maintaining metals 
 other than copper in the state of oxide in the slag. 
 
 OLD COPPER. For turnings, waste of workshops, &c., 
 yielding 97 to 98 per cent, by the assay, and containing, 
 in fact, not much foreign matter except a little mixed dust 
 
ENGLISH COPPER ASSAY. 
 
 453 
 
 or dirt, we take care first to glaze the crucible by fusing 
 in it a little borax and nitre ; then we treat the matters by 
 a simple washing, the slags of which we work for prill. 
 This last is often very considerable. 
 
 SECT. IV. SUMMARY CONSIDERATIONS COMPARISON OF THE 
 EESULTS WITH THE ANALYSIS BY THE WET WAY. 
 
 After this detailed account of the numerous operations 
 which the metal undergoes before attaining the state of 
 button and prill, it would be superfluous to insist upon the 
 practical difficulty of the Cornish method. 
 
 Nevertheless, in experienced hands, and in the case of 
 daily practice, it is still a rapid method, allowing us to 
 treat almost uniformly the different varieties of copper 
 mineral, and at the least to remedy during the operation 
 itself the impurities which show themselves. 
 
 As to the metallurgic accuracy, below is a small table 
 showing comparatively the produce by the dry way (es- 
 timated by a Cornish assayer) and that which we have 
 obtained by the most precise methods of the wet way. It 
 comprehends six samples, whose richness varies within 
 sufficiently great limits. 
 
 
 
 
 
 Nature of the sample 
 
 Dry way 
 
 Wet way 
 
 Difference 
 
 and produce 
 
 D. 
 
 W 
 
 W-D 
 
 Regulus of Chili ..... 
 
 564 = 56-250 
 
 58-40 
 
 2-150 
 
 Green copper carbonate of Castile . 
 
 9f = 9-750 
 
 11-52 
 
 1-770 
 
 Variegated copper, Huel Damsel . 
 Pyrites, West Wheal Seton . . 
 
 10i = 10-500 
 8|= 8-375 
 
 11-30 
 8-40 
 
 0-800 
 0-025 
 
 United Mines .... 
 
 8 = 8-000 
 
 10-38 
 
 2-380 
 
 Devon Great Consols 
 
 4|= 4-625 
 
 5-60 
 
 0-975 
 
 
 S(W-D) . 
 
 . 
 
 8-100 
 
 
 6 
 
 
 1-350^ 
 
 By adding the result given by the last five minerals 
 we find 
 
 = 9-44 
 b 
 
 By taking the ratio 
 
454 ENGLISH COPPER ASSAY. 
 
 we see that we must add to the richness indicated by the 
 Cornish assay about f of that result, and by taking the 
 ratio 
 
 that the loss is -J of the copper, if we consider a mineral 
 of 9 or 10 per cent. 
 
 Without wishing to draw a conclusion altogether 
 general from so small a number of analyses, we neverthe- 
 less think they suffice to show that the Cornish method 
 occasions losses always sensible and sometimes considerable. 
 We think we may assert that upon the whole of the 
 Cornish minerals whose mean richness varies from 6 to 7 
 per cent., the loss by the assay is not less than 20 per 
 cent, of the contained copper, and that for certain pyrites 
 of 3 to 4 per cent, it attains 30 and 40 per cent, of the 
 metal. 
 
 The principal causes of these losses are (1st) the 
 quantity more or less great of copper left in the slag of 
 the regulus ; (2nd) and especially the carrying away of 
 copper by the vapours of common salt in the fusion for 
 coarse copper, the washing or washings, the refining and 
 the treatment of the slag for prill. 
 
 In consequence we think they ought to bear princi- 
 pally on the oxidised minerals for which we make a rich 
 regulus, and still more on the impure minerals, which 
 besides a rich regulus have undergone several washings. 
 Thus the minerals of Algeria, grey copper, assayed some 
 years ago at the School of Mines, have given a produce 
 much higher than that indicated by the Cornish assayers. 
 
 It may be said, therefore, that the results of the Cornish 
 assay do not fall short of the truth by a fixed quantity,, 
 but become more and more inaccurate the poorer the ore. 
 The smelter gets out of the ore more metal than the assay 
 indicates. 
 
 The most frequent alloys of copper, i.e. brass, German 
 silver, gun-metal, &c., cannot be assayed in a reliable 
 manner in the dry way ; German silver because the 
 
ENGLISH COPPER ASSAY. 455 
 
 nickel could not be removed at all, or only with great 
 difficulty, and the rest because zinc and tin give such diffi- 
 cultly fusible oxides that they could not be properly 
 removed in the refining. 
 
 In the alloys of copper with silver -, gold, and platinum, 
 the copper may be estimated from, the loss arising from 
 cupellation with lead. 
 
 In all assays of copper in the dry way the silver or 
 auriferous silver contained in the assay sample cannot be 
 removed, and it is generally pretty completely collected 
 in the copper obtained. These copper assays give no- 
 where any indications whether gold or silver is present 
 or not ; and the amount of these metals which may be 
 present must therefore be both sought for and estimated 
 by a special assay for them. If they are found, and in 
 sufficiently large quantity, they are deducted from the 
 weight of the copper. 
 
 The dry assay is mostly found in practice in smelting 
 works, where, even in the hands of less scientifically edu- 
 cated than skilful assayers, with the character of the assay 
 substance once known, and suitable practice in following 
 out the separate manipulations, it gives results which 
 suffice for the business of working copper in the large 
 way. 
 
 b. For Ores and Products of Class II. 
 
 THE LAKE SUPERIOR FIRE ASSAY. 
 
 We are indebted to Dr. E. D. Peters's work on 
 ' Modern American Methods of Copper Smelting ' for the 
 following excellent description of the dry method of assay 
 in use in America.* The ordinary English fire-assay is 
 little suited to American conditions. The Lake Superior 
 fire-assay, on the contrary, is not only quick and inexpen- 
 sive, but compares favourably in accuracy with the best 
 
 * In his description of this mode of assay Dr. Peters acknowledges his 
 indebtedness to Mr. Maurice B. Patch of Houghton, Michigan, for valuable 
 assistance in the preparation of this section on the Lake Superior assay. The 
 position held by Mr. Patch as chemist to the Detroit and Lake Superior 
 Copper Company is a sufficient guarantee of the accuracy of the following 
 description. 
 
456 AMERICAN COPPER ASSAY. 
 
 wet methods. It is so peculiarly adapted to the conditions 
 that have given it birth, that no American work on the 
 metallurgy of copper would be complete without a detailed 
 account of it, especially as our docimastic literature up to 
 this time has made little mention of it. In the Swansea 
 assay, the substance under treatment consists usually of a 
 mixture of sulphides and gangue rock, which necessitates 
 a series of calcinations and fusions, culminating in a button 
 of impure copper, which has still to be refined at a con- 
 siderable loss. The Lake Superior assay er has the simpler 
 problem of dealing only with native or oxidised compounds 
 of copper that can be reduced to the metallic state at so 
 low a temperature as to preclude the adulteration of the 
 copper button with any other metallic substances, and 
 thus obviate the necessity of any refining process. In 
 spite of the apparent simplicity of this method, it demands 
 a good deal of skill and experience to obtain correct re- 
 sults ; but these once acquired, no assay can excel it in 
 accuracy and celerity. 
 
 A glance at the composition of the substances operated 
 on will render clear the objects to be accomplished. The 
 material assayed consists of the concentrates from the jigs, 
 tables, buddies, and other concentrating machines. This 
 material is technically termed ' mineral,' and varies greatly 
 in richness, composition, and size of particles, ranging in 
 copper from 10 to 97 percent., and in some instances con- 
 taining a gangue of nearly pure ferric oxide, while in 
 others it is highly silicious. Nearly all grades of mineral 
 contain a considerable proportion (from 3 to 10 per cent.) 
 of metallic iron from the stamp heads, while a sample 
 containing 50 per cent, of titanic iron-sand is no unusual 
 occurrence. It can readily be seen that no small skill is 
 required so to flux these various mixtures as to obtain a 
 clean and fusible slag, and a button of copper free from 
 iron or other metals that may be reduced with compara- 
 tive ease, and thus yield a far too high result. 
 
 Sampling. The mineral is received from the mines, 
 packed in strong barrels, weighing in the damp condition 
 in which it arrives from 500 to 2,000 pounds, its weight 
 
AMERICAN COPPER ASSAY. 457 
 
 depending on its degree of concentration, the character of 
 its accompanying gangue, &c. As this material is to be 
 refined at once, the barrels are emptied on the iron plates 
 that form the floor in the neighbourhood of the charging- 
 door of the refining-furnace. After the contents of each 
 barrel have been thoroughly and separately mixed, a small 
 sample is taken from every package and put into a tightly 
 covered copper can. Only the samples from casks of the 
 same grade of mineral are placed in any one can, as each 
 quality is assayed by itself, although six or more different 
 grades of mineral may go to make up the sixteen barrels 
 that usually form a furnace charge. If two or more furnaces 
 are simultaneously in operation, the samples of the same 
 grade are mixed together, to avoid the unnecessary multi- 
 plication of assays ; the cans containing a little water in 
 the bottom, into which the tight copper cans are set, to 
 prevent any loss of moisture in the sample, which might 
 occur despite the close cover. 
 
 Fluxes. Sodium bicarbonate, borax, potassium bitar- 
 trate (cream of tartar), ferric oxides, sand, and slag from 
 the same operation are used to flux the gangue and other 
 worthless constituents, and effect the proper reduction of 
 the copper. The chief impurity to be dreaded is sulphur, 
 for which reason the best quality of sodium bicarbonate 
 must be purchased, and potassium bitartrate must be used 
 instead of argols. The borax and soda are prepared by 
 being melted in iron ladles, to drive off their water of 
 crystallisation, and then pulverised through a twenty- 
 mesh screen ; clean, well-fused slag from former operations 
 is reduced to the same degree of fineness, while. the oxide 
 of iron flux is prepared by pulverising selected fragments 
 of specular iron through a fifty-mesh sieve. Any clear 
 quartz sand answers for the silica needed. 
 
 Furnace. A common natural draught melting-furnace 
 is used, an inside measurement of 9-J by 18 inches being 
 large enough to accommodate six Hessian crucibles. These 
 are set in rows of three on two thin fire-bricks, the latter 
 resting on the longitudinal grate-bars, and serving to raise 
 the crucibles to the zone of greatest heat. Soft coal, broken 
 
458 AMERICAN COPPER ASSAY. 
 
 to egg size, forms the customary fuel, and is carefully filled 
 in around the charged crucibles, which are not placed in 
 the furnace until the latter is in full heat. The crucibles 
 employed are four inches high and three inches in diameter, 
 and are provided with well-fitting covers made at the works 
 from a mixture of fireclay and sand ; these are the more 
 necessary because the assay often fills the crucible to 
 within half an inch of the top. 
 
 The skill of the assayer is nowhere more evident than 
 in the fluxing of the different grades of mineral, the com- 
 position of which was briefly noticed in the opening para- 
 graph of this chapter. It is, of course, familiar to all 
 chemists that sodium bicarbonate and ferric oxide act as 
 powerful bases, while the electro-negative elements are 
 represented by borax and sand : the potassium bitartrate 
 exercises a strong reducing action, as well as furnishing 
 an active base. The slag equalises the entire mixture, 
 being capable of neutralising a considerable amount of 
 either base or acid, and it covers the molten metal and 
 protects it from oxidation. It is not to his skill in fluxing 
 alone that the assayer trusts ; of almost equal importance 
 are the degree of temperature maintained and the length 
 of time that the assays are left in the furnace. 
 
 Good results can only be obtained by shortening the 
 period of fusion to the utmost. This demands a very hot 
 furnace at the outset, good fuel, and a lively draught. 
 Under these conditions an easily fusible assay will pro- 
 bably be entirely finished in 20 minutes, while from 25 
 to 30 minutes are required for different samples. It is 
 quite safe to assert that if the time necessary for a perfect 
 fusion is increased to 40 minutes, the resulting button 
 will contain sufficient impurities reduced from the slag to 
 give a result from 2-|- to 6 per cent, too high. 
 
 This assay is applicable to silicates as well as oxides and 
 native copper, and the results obtained from the assay of 
 both refining and blast furnace slags cannot be excelled 
 in accuracy by any other method. 
 
 A table of the different weights of fluxes used in assay- 
 ing the various grades of mineral from the Peninsular 
 
AMERICAN COFFEE ASSAY. 
 
 459 
 
 Copper Company's works is annexed, as well as the mix- 
 ture adopted for reverberatory slags for very silicious 
 
 ores. 
 
 Minerals 
 
 Weight, 
 grains 
 
 Borax, 
 grains 
 
 Soda, 
 grains 
 
 Slag, 
 grains 
 
 Potassium 
 bitartrate, 
 grains 
 
 Sand, 
 grains 
 
 Iron ore, 
 grains 
 
 Xo. 
 
 Per cent. 
 
 
 
 
 
 
 
 
 copper 
 
 
 1 
 
 
 
 
 1 
 
 92 
 
 1,000 
 
 60 
 
 55 200 
 
 300 
 
 
 
 
 
 2 
 
 86 
 
 1,000 
 
 60 
 
 60 
 
 180 
 
 300 
 
 
 
 
 
 3 
 
 60 
 
 500 
 
 100 
 
 80 
 
 
 
 300 
 
 
 
 
 
 4 
 
 83 
 
 500 
 
 150 
 
 160 
 
 
 
 300 
 
 150 
 
 
 
 5 
 
 20 
 
 500 
 
 190 
 
 200 
 
 
 
 300 
 
 175 
 
 
 
 # 
 
 35 
 
 500 
 
 140 
 
 140 
 
 
 
 300 
 
 
 
 100 
 
 t 
 
 5 to 20 
 
 500 
 
 200 
 
 200 
 
 
 
 300 
 
 
 
 
 
 The percentage of slag-forming materials being so 
 small in Nos. 1 and 2, it requires but a slight amount of 
 borax and soda to flux them, while an addition of neutral 
 slag is necessary to protect the molten copper. A smaller 
 quantity of the ore is weighed out in the succeeding assays, 
 as they are so poor in copper that a large amount of flux 
 is required by the great quantity of gangue, so that the 
 capacity of the ordinary crucibles would be greatly ex- 
 ceeded if 1,000 grains were used. No. 3 mineral contains 
 just sufficient ferric oxide to form a good slag with the 
 mixture given ; while in Nos. 4 and 5 this substance, as 
 well as metallic iron, increases to such an extent as to 
 require the addition of a considerable proportion of sand 
 to flux this base and to prevent the adulteration of the 
 button with metallic iron. 
 
 The sample of Calumet and Hecla tail-house mineral 
 given is typical of the treatment of very silicious material. 
 There is nothing remarkable in the considerable proportion 
 of borax (an acid flux) used with even highly quartzose 
 ores ; for, in addition to the fluxing powers of the soda that 
 it contains, a boro-silicate is very much more fusible than a 
 simple silicate. No peculiarities exist in the execution 
 of this assay ; the ore and fluxes are thoroughly mixed on 
 glazed paper and covered with a thin layer of potassium 
 bitartrate after being poured into the crucible. In the 
 
 * Calumet and Hecla tail-house mineral. t Rich slag from refining. 
 
460 AMERICAN COPPER ASSAY. 
 
 No. 1 mineral, which is nearly as coarse as split peas, 
 fragments of iron frequently exist, which come from the 
 stamp heads, and must be picked out of the sample after 
 weighing out for assay : not that cast iron will alloy with 
 copper, but that the fragments will be found imbedded in 
 the copper button after cooling. 
 
 The results obtained by this method are surprisingly 
 accurate. Duplicate estimations of the lower grade 
 samples seldom vary more than 0*1 or 0'2. A difference 
 of 0*4 per cent, is a rare occurrence even in the higher 
 classes of mineral, where the size of the metallic fragments 
 renders the sampling, and even the weighing out, of a 
 correct assay a matter of some uncertainty. 
 
 A few results from Mr. Patch's notes will confirm 
 these statements. An average series of tests on cupola 
 slags by the colorimetric method for the period of a month, 
 duplicated by the fire assay, gave a result 0-05 per cent, 
 lower for the latter test, the slag containing about 0*5 of 
 one per cent. 
 
 As an illustration of the results of this system when 
 applied to very rich ore, a comparative test was made for 
 eight days on No. 1 Calumet and Hecla mineral, with the 
 following results : 
 
 Battery assay .... 89-100 per cent. 
 Fire assay . . . 88-812 
 
 A similar test on No. 2 Calumet and Hecla mineral : - 
 
 Battery assay .... 77*590 per cent. 
 Fire assay . 77'657 
 
 A similar test with various samples : 
 
 No. Battery assay Fir 
 
 r89-501 
 
 >aa.ftn 
 mean = 89-544 ^ . >mean = 89*92 
 oy-/u 
 
 L 89-70 I 
 
 f 77-40] 
 
 > mean - 77-740 J ^'.^ >mean = 77'50 
 
 L77-40J 
 
 It is a somewhat curious fact that the slight loss of 
 about 0-25 per cent, of copper which results from the 
 passage of a minute portion of the metal into the slag 
 
ASSAYS IX THE WET WAY. 461 
 
 is just about counterbalanced by the impurities in the 
 copper button from the reduction of ferric oxide, the 
 amount of which is indicated by the following analyses 
 of copper buttons the only weighable impurity being 
 iron : 
 
 Copper, per cent. Copper, per cent. Copper, per cent. 
 
 09-83 99-76 99-51 
 
 99-84 99-80 99-87 
 
 99-52 99-46 99-79 
 
 This account of a little-known process will doubtless 
 remove the impression sometimes held by chemists, that 
 the Lake Superior copper assay is a clumsy and imperfect 
 operation, and unworthy any advanced system of metal- 
 lurgy. 
 
 B. ASSAYS ix THE WET WAY. 
 a. Colorimetric Copper Assays. 
 
 These are based upon the fact that ammonia, added 
 in excess to the solutions of salts of copper, produces a 
 beautiful azure-blue colour, whose intensity depends upon 
 the quantity of copper dissolved. By comparing the 
 shades of blue colour in equally thick layers of the dis- 
 solved ammoniacal assay substance (assay fluid) with a 
 normal or standard ammoniacal fluid whose copper 
 contents are known, the quantity of copper in the former 
 can be calculated when its volume is measured. 
 
 To Heine, the superintendent of the smelting works in 
 Mansfield, belongs the merit of having first successfully 
 employed this reaction for the estimation of small per- 
 centages of copper ; and later it has been also extended 
 by Jacquelain, Von Hubert, and Muller, to the estimation 
 of larger quantities of copper. 
 
 1. HEIXE'S COLOEIMETRIC METHOD. 
 
 For the estimation of the quantity of copper in bodies 
 poor in this metal, e.g. in slags, lead matte, litharge, 
 crude lead, and 'other plumbiferous metallurgical pro- 
 ducts, tin, cupelled silver, &c. in short, in all substances 
 
462 THE ASSAY OF COPPER. 
 
 which contain from a trace to about 1 per cent, or a little 
 more of copper, this method is the most advantageous to 
 be used. 
 
 After the assay sample has been reduced to as fine a 
 state of mechanical subdivision as possible, which with 
 slags is best attained by sifting or washing them, one 
 centner (3-4 grammes) of it is weighed out and dissolved, 
 or so completely decomposed by a suitable acid that in 
 the residue, which is to be filtered and well washed, no 
 more copper remains behind. For this purpose nitric acid 
 or aqua regia is employed, according to the character and 
 particular behaviour of the substance, and the nitric acid 
 is concentrated or somewhat diluted, as may be required. 
 The solution is either immediately, or after the copper 
 has been first precipitated by sulphuretted hydrogen gas 
 and again dissolved, strongly supersaturated with caustic 
 ammonia, and the precipitate, if any, thereby produced, 
 digested in caustic ammonia for a considerable time, with 
 frequent stirring at a very gentle heat (30-40 G.), then 
 filtered off and thoroughly washed. According to the 
 quantity of copper present, and according to the degree 
 of dilution, the solution obtained will appear more or 
 less strongly coloured blue. The volume of the solution 
 is measured in graduated vessels, and the intensity of the 
 colour compared with and estimated from fluids, which 
 have been previously prepared as standard fluids, and 
 which for a definite volume contain a definite, accurately 
 weighed quantity of copper, that has been dissolved in 
 nitric acid, precipitated by caustic ammonia, and redis- 
 solved in excess of the same. From the measured volume, 
 and the intensity found by comparison, the quantity of 
 copper is then found by calculation. 
 
 Heine proposes standard fluids with one, two, three, 
 and four assay loth of copper in one ounce (two loth, 
 commercial weight) of the ammoniacal fluid. These four 
 standard fluids are all-sufficient. 
 
 If the French weights and measures are used, standard 
 fluids are taken with -001, -002, -003, -004 gramme of 
 copper to every twenty -five cubic centimetres of the fluid. 
 
COLORIMETRIC ASSAYS. 463 
 
 The graduated vessels (cylinders) required for the 
 preparation of the standard fluids, as well as for the 
 measuring of the assay fluid, can be easily prepared by 
 the assay er himself. One quarter of an ounce of water is 
 weighed out a number of times in succession and poured 
 into the cylinder, and each time the height of the fluid is 
 marked in a durable manner on the glass with a diamond, 
 or by etching it with hydrofluoric acid vapour, &c. Also 
 earthen or porcelain measures, that are prepared and 
 marked for the volumes that hold one, two, three, four, 
 &c., ounces of water, may be used. 
 
 It is not practicable to replace the volumetric measure- 
 ment by weighing, for the quality and quantity of those 
 substances which are soluble in acids and not precipitated 
 by ammonia, or are again dissolved by it, may vary greatly 
 in the assay. 
 
 In the formation of the normal fluids, two assay pounds 
 of chemically pure (galvanic) copper are weighed out on a 
 good balance, dissolved in nitric acid, the solution super- 
 saturated with caustic ammonia, and placed in a graduated 
 cylinder, which is divided to whole, half, and quarter 
 ounce volumes of water, and then water enough is added 
 to bring the fluid to the sixteen-ounce mark. The fluid 
 then contains -f-g-=4 loth of copper per ounce. Six ounces 
 of this four-loth solution are then taken, two ounces of 
 water added to it, and eight ounces of fluid obtained, with 
 ^ = 3 loth of copper to one ounce of water. The two-loth 
 solution is formed in a similar way by diluting four ounces 
 of the four-loth solution to eight ounces ; the one-loth, by 
 diluting four ounces of the four-loth normal fluid to six- 
 teen ounces. In the measuring of the assay fluid it is 
 estimated within one-eighth of an ounce, which is suffi- 
 ciently close. If in the dilutions a mistake is actually 
 made of one-sixteenth of an ounce, the maximum of possi- 
 bility, the error amounts to about two cubic centimetres, 
 which in a whole mass of fluid of 200-500 cubic centi- 
 metres has no influence upon the solution that can be 
 detected with the eye. 
 
 The preservation of the standard fluids, as well as the 
 
464 THE ASSAY OF COPPER. 
 
 comparison of the blue assay fluids with them, must take 
 place in glass vessels closed with ground-glass stoppers. 
 These vessels must have the same form and size, consist of 
 the same colourless glass, and have an equal thickness of 
 glass in the smooth side walls. The last condition is 
 obtained the surest by grinding. This grinding, however, 
 which notably increases the cost of the glasses, is not in- 
 dispensably necessary if the vessels are carefully formed 
 and blown in a good glass-house. An oblong form is most 
 advantageous for the vessels. They hold about an ounce 
 and a half of fluid, and are about two inches long, two and 
 a half inches high, and one inch wide, with walls about 
 one-eighth of an inch thick. 
 
 The glasses are very advantageously formed from 
 an unblemished sheet of plate glass of equal thickness 
 throughout, by cementing the sides together and the 
 insertion of a glass neck. The assayer has in the form of 
 vessel indicated a triple control in the comparison of the 
 assay fluid with the normal solution according as he looks 
 through the fluid in three different directions. 
 
 The digestion of the assay sample with acid may take 
 place in any suitable vessel whatever a glass flask, a 
 beaker covered with a watch-glass, &c., only no thumping 
 and spirting of the fluid should be possible in the process. 
 The nitric acid, &c., must be added little by little. The 
 time required for this may vary greatly. The solution of 
 cupelled silver* skimmings , &c., with nitric acid is finished 
 in a short time ; on the other hand, in the examination 
 of difficultly decomposable slags, with which concentrated 
 nitric acid or aqua regia will always be used, the digestion 
 often requires to be continued at a warm temperature for 
 two to three times in twenty-four hours. The mass must 
 be frequently stirred with a glass rod, because many slags 
 decompose rapidly with evolution of heat, form a thick 
 jelly, and deposit a crust on the bottom of the glass sub-, 
 mono-, and bi-silicate slags mostly decompose readily, 
 
 * With cupelled silver, after dissolving in nitric acid, the silver may be 
 precipitated with sodium chloride, the silver chloride filtered, washed, and the 
 solution then mixed with caustic ammonia. 
 
COLORIMETRIC ASSAYS. 465 
 
 higher silicates resist complete decomposition by aqua 
 regia and then a preliminary solvent ignition or fusion 
 with potassium carbonate or calcined sodium carbonate, 
 or, better, a mixture of both, is, necessary, precisely in the 
 manner given in the wet assay of copper. Here also it 
 does no harm if some of the substance of the crucible 
 remains adhering to it. 
 
 The decomposition of the slags by acid is complete 
 when in the stirring with a glass rod no more grating can 
 be perceived. 
 
 After hot water has been added to the decomposed 
 assay, the residue is collected on a filter, well washed out, 
 without diluting the filtrate too largely, and the copper 
 precipitated from the solution, if necessary, with sulphu- 
 retted hydrogen gas, especially when a notable quantity of 
 alumina and iron is present, whose slimy precipitates from 
 the immediate precipitation with ammonia always retain 
 copper. This precipitation of the copper has also the 
 advantage that, as cobalt and nickel do not precipitate 
 with it, the colouring effects which they would produce, 
 if present, are removed. Since the copper sulphide re- 
 quires for its solution but a few drops of nitric acid, in 
 the succeeding treatment of the solution with ammonia, 
 but a small quantity of ammoniacal salt is formed, and the 
 specific gravity of the coloured fluid varies but very little 
 from that of water and the normal solution. With the 
 increase of the specific gravity of the assay solution, its 
 volume is considerably increased, and therefore it gives 
 too large a measure in the direct precipitation with am- 
 monia. If the precipitation with sulphuretted hydrogen 
 gas is completed in four to six hours, the copper sulphide 
 is filtered out, thoroughly washed with cold water contain- 
 ing sulphuretted hydrogen, the filter dried, ignited in a 
 porce 7 Crucible, the copper oxide formed warmed with 
 a few dro of nitric acid or aqua regia, supersaturated 
 with ainm< .a, filtered, and well washed, till the washings 
 are no longer tinged bluish. 
 
 A precipitation of the copper with iron wire, from a 
 solution evaporated with sulphuric acid, and a re-solution 
 
 H H 
 
466 THE ASSAY OF COPPER. 
 
 of the copper in nitric acid, consumes less time. If the 
 copper is not precisely precipitated, errors of some 30 
 per cent, and more of the whole amount of copper may 
 occur. By repeated solution of the iron precipitate and 
 precipitation with ammonia, all the copper cannot, how- 
 ever, be extracted. 
 
 In the examination of litharge, the solution in nitric 
 acid may be dispensed with. The copper oxide can be at 
 once extracted from it with caustic ammonia ; however, 
 the litharge and ammonia must then be allowed to work 
 at least twenty-four hours on each other, with very diligent 
 stirring, and, moreover, the litharge must be ground very 
 fine. 
 
 The ammoniacal solution obtained from the assay is 
 now well stirred, so that it may mix with perfect unifor- 
 mity with the last washings ; then, either the whole, or a 
 part of it, is placed in a clean assay glass, and compared 
 with the standard fluids in similarly formed glass vessels 
 standing on a sheet of white paper. Should it correspond 
 with none of them in the intensity of its colour, the whole 
 of the fluid is diluted somewhat with water, until this is 
 the case. Its volume is thereupon measured in the glass 
 vessel graduated in ounces, &c., and noted. For a check, 
 the dilution may be carried still farther till the colour of 
 the assay corresponds to the next more faintly coloured 
 standard fluid, and then the increased volume be measured 
 anew. This might perhaps be still again repeated, but it 
 becomes more and more uncertain. The calculation of 
 the percentage of copper, from the intensity and the volume 
 found, then presents no further difficulty. 
 
 Suppose that the assay fluid agrees with the normal so- 
 lution of four loth of copper to the ounce of water, and 
 its quantity amounts to five ounces, then the quantity of 
 copper in the centner of the assay substance is 5 x 4=20 
 loth. This fluid further diluted till it equals the normal 
 solution with three loth of copper, must measure six and 
 two- third ounces if the obtained value of twenty loth is to 
 be confirmed. 
 
 According to Heine's experiments, the possible error 
 
COLOEIMETEIC ASSAYS. 467 
 
 of observations in the comparisons and measurement de- 
 scribed amounts as a maximum with the stronger normal 
 solutions (with sixteen loth and over) to from three-quarters 
 to one loth, with the weaker ones to scarcely half a loth 
 of copper. In a centner of the assay substance, one loth 
 of copper '03 per cent, can still be estimated with cer- 
 tainty. 
 
 Le Play's Method. Le Play estimated in finely pul- 
 verised and carefully washed copper slags, the copper in 
 one gramme of the poorest slags to within half a milli- 
 gramme, and of the richest slags to within one milli- 
 gramme, by using twenty-six standard fluids with various 
 percentages of copper in cylindrical vessels. The com- 
 parison of colours in round vessels is more uncertain than 
 in oblong ones, since in the former the light is dissipated 
 and shadows are produced. 
 
 T. 0. Cloud's Method. According to Mr. T, 0. Cloud 
 the copper cannot be completely extracted from the finest 
 ground slags, even after three days' digestion with aqua 
 regia, and subsequent treatment with sulphuric acid. He 
 recommends fusing the slag with four parts of mixed 
 potassium and sodium carbonates and J part potassium 
 nitrate. The fused mass is extracted with dilute sul- 
 phuric acid, the liquid evaporated down, and the copper 
 estimated galvanically or colorimetrically. 
 
 Endemanris Method. As a delicate test for small quan- 
 tities of copper, Dr. Endemann (' Annalen der Chernie ') 
 adds to the dilute solution concentrated hydrobromic 
 acid, when a dark brownish-red or violet colour is at once 
 produced. 
 
 This reaction is so delicate that T -^Q- milligramme of 
 copper can be detected with certainty. One drop of a 
 solution containing this small quantity of copper is brought 
 on a watch-glass, then one drop of hydrobromic acid is 
 added, and the solution is then allowed to evaporate slowly 
 by standing the glass in a warm place. When the whole 
 has been concentrated to about one drop, this will distinctly 
 show a rose-red colour. The colour thus produced is 
 about three or four times as distinct as the one which is 
 
 H TT 2 
 
408 THE ASSAY OF COPPEE. 
 
 obtained by the addition of potassium ferrocyanide. Of 
 other metals which are examined in this direction, we find 
 only iron to be apt to interfere with this reaction, and then 
 only when it is present in considerable quantity. 
 
 This reaction may also be utilised as a colorimetric 
 test for the quantitative estimation of small quantities of 
 copper. 
 
 If a substance contains so little copper that the fluid 
 does not equal the most faintly coloured standard fluid in 
 intensity of colour, the assayer must endeavour to remedy 
 the matter by evaporating till this is the case. An evapora- 
 tion is, however, to be avoided if possible, first because of 
 the loss of time, and also because other precipitates, carbo- 
 nate of lime, &c., are apt to be caused by it, and because, 
 when it has to be continued too long, so much ammonia is 
 apt to be volatilised, that a new addition of it becomes 
 necessary. 
 
 This method of assaying soon finds the limits of its 
 accuracy in an increasing percentage of copper in the 
 assay sample, since with fluids rich in copper, and there- 
 fore strongly coloured blue, the errors of observation soon 
 amount to several loth. And to seek then to better one- 
 self by diluting largely, yields no more accurate results^ 
 since a small error of observation in estimating the 
 intensity of the colour, is so much the more multiplied in 
 the calculation of the value by the greater number of the 
 ounces. 
 
 If nickel is contained in the assay substance, the assay 
 cannot be conducted in the way prescribed, since the 
 nickel is extracted by the acids, and dissolves also in 
 caustic ammonia with a blue colour. The assay may also 
 become uncertain from the presence of much m,anganese, 
 cobalt, or chromium, since they render the hue of the blue 
 colour dingy. Chromium may be completely removed by 
 a slight boiling of the ammoniacal fluid ; not so cobalt. 
 The presence of vanadium or molybdenum does no harm. 
 
 If nickel, or much cobalt and manganese, is contained 
 in the assay substance, the solution obtained by acids and 
 filtered, though not further diluted, must first be decom- 
 
COLOKIMETE1C ASSAYS. 
 
 posed by metallic iron. What is thrown down by the iron 
 is collected on a small filter, washed thoroughly, and then, 
 together with the filter, treated with dilute nitric acid. 
 When the copper is all dissolved, this solution is supersatu- 
 rated with caustic ammonia and then managed as above. 
 With higher percentages of copper the process of the 
 Swedish copper assay is used for estimating the value. 
 The precipitation of the copper may also be performed 
 with sulphuretted hydrogen gas. 
 
 Le Play removes the injurious influence of manganese, 
 nickel, and cobalt, by allowing the green or violet-coloured 
 ammoniacal solution to stand open to the air for some 
 time in a moderately warmed drying furnace, whereby a 
 few variously coloured gelatinous flocks are gradually 
 deposited, and the fluid, after the addition of a few drops 
 of ammonia, then becomes pure blue. 
 
 According to Jacquelain and Von Hubert, nickel and 
 cobalt are in a simple way rendered perfectly harmless by 
 gradually adding white pulverised marble to the solution 
 of the assay substance, until the effervescence ceases, and 
 then warming the whole on the sand-bath, whereby all the 
 copper is perfectly precipitated as carbonate, while nickel 
 and cobalt remain dissolved. It is now filtered, washed, 
 the residue dissolved in nitric acid, and the solution treated 
 as already explained, with ammonia. By the addition of 
 potassium carbonate to the ammoniacal fluid, and heating, 
 all the manganese precipitates, while the copper remains 
 dissolved in the excess of ammonia, and can be separated 
 from the manganese precipitate by filtration. The manga- 
 nese must have been present as oxide in the original solu- 
 tion in order that the precipitation by potassium carbonate 
 may be perfect. 
 
 The assayer may convince himself whether nickel 
 or cobalt is present, by slightly supersaturating a blue 
 ammoniacal solution^ obtained by the ordinary process of 
 assaying, with hydrochloric or sulphuric acid, then pre- 
 cipitating the copper completely with iron, filtering the 
 residual solution, concentrating somewhat, if necessary, 
 and then supersaturating with ammonia. If the fluid 
 
470 THE ASSAY OF COPPEK. 
 
 remains colourless, neither of the two metals is present : a 
 blue colour indicates nickel, a red one cobalt. 
 
 Sometimes the normal solutions, which when freshly 
 prepared appear azure blue, assume a greenish hue, which 
 renders the comparison difficult, if not impossible. Copper 
 nitrate produces with ammonia a pure azure blue, copper 
 sulphate a lilac colour, and copper chloride greenish hues. 
 Sulphuric and hydrochloric acid are therefore to be avoided 
 as much as possible in the solution. But, nevertheless, an 
 assay fluid may sometimes e.g. by standing some time in 
 the air, or by slow filtration become green, in which case 
 the colour is destroyed by a few drops of nitric acid, and 
 ammonia added anew. But sometimes also the greenish 
 colour disappears, if the solution stands in a covered vessel 
 in the air, or by the addition of a few drops of red cobalt 
 ammonio-oxide. 
 
 According to Mliller, also, the colour stands in the 
 closest connection with the quantity of ammonia employed, 
 and it therefore leads to greater accuracy in the assay if- a 
 titrated solution of ammonia is used, and the volume of 
 the ammoniacal fluid noted, which, after neutralisation of 
 the residual free acid, is used for the solution of copper. 
 The solution appears more intense when viewed with a 
 grey background than with a white one. A greenish blue 
 colouring becomes the more noticeable the greater is the 
 excess of ammonia, or the more ammoniacal salts are in 
 the solution. 
 
 Camelly's Method. Dr. T. Camelry (Manchester Philo- 
 sophical Society) gives the following colorimetric method 
 for estimating small quantities of copper : 
 
 The method of analysis consists in the comparison of 
 the purple-brown colours produced by adding to a solu- 
 tion of potassium ferrocyanide first, a solution of copper 
 of known strength, and secondly, the solution in which 
 the copper is to be estimated. 
 
 The solution arid materials required are as follows : 
 
 (1) Standard Copper Solution. Prepared by dissolv- 
 ing 0-393 grammes of pure CuS0 4 .5H 2 in one litre of 
 water, 1 c.c. is then equivalent to 0-1 mgrm. Cu. 
 
COLORIMETRIC ASSAYS. 471 
 
 (2) Solution of Ammonium Nitrate. Made by dissolv- 
 ing 100 grms. of the salt in one litre of water. 
 
 (3) Potassium Ferrocyanide Solution. Containing one 
 part of the salt in 25 parts of water. 
 
 (4) Two glass cylinders, holding rather more than 150 
 c.c. each, the point equivalent to that volume being marked 
 on the glass. They must, of course, both be of the same 
 tint and as nearly colourless as possible. 
 
 (5) A burette, marked T L c.c., for the copper solution, 
 a 5 c.c. pipette for the ammonium nitrate, and a small 
 tube to deliver the potassium ferrocyanide in drops. 
 
 The following is the method of analysis : Five drops 
 of the potassium ferrocyanide are placed in each cylinder, 
 and then a measured quantity of the neutral solution in 
 which the copper is to be estimated into one of them 
 (A), and both filled up to the mark with distilled water, 5 
 c.c. of the ammonium nitrate solution added to each, and 
 then the standard copper solution runs gradually into (B), 
 till the colours in both cylinders are of the same depth, the 
 liquid being well stirred after each addition. The number 
 of cubic centimetres used is then read off. Each cubic 
 centimetre corresponds to 0*1 mgrm. of copper, from 
 which the amount of copper in the solution in question 
 can be calculated. 
 
 The solution in which the copper is to be estimated 
 must be neutral, for if it contains free acid the latter lessens 
 the depth of colour, and changes it from a purple-brown 
 to an earthy brown. If it should be acid it is rendered 
 slightly alkaline with ammonia, and the excess got rid 
 of by boiling. The solution must not be alkaline, as the 
 brown coloration is soluble in ammonia and decomposed 
 by potash ; if it is alkaline from ammonia this is remedied 
 as before by boiling it off, while free potash, should it be 
 present, is neutralised by an acid and the latter by am- 
 monia. 
 
 Lead when present in not too large quantity has little 
 or no effect on the accuracy of the method. The precipi- 
 tate obtained on adding potassium ferrocyanide to a lead 
 salt is white, and this, except when present in compara- 
 
472 THE ASSAY OF COPPER. 
 
 tively large quantity with respect to the copper, does riot 
 interfere with the comparison of the colours. 
 
 When copper is to be estimated in a solution contain- 
 ing iron the following is the method of procedure to be 
 adopted. To the solution a few drops of nitric acid are 
 added in order to oxidise the iron, the liquid evaporated to 
 a small bulk, and the iron precipitated by ammonia. Even 
 when very small quantities of iron are present this can be 
 done easily and completely if there is only a very small 
 quantity of fluid. The precipitate of ferric oxide is then 
 filtered off, washed once, dissolved in nitric acid, and repre- 
 cipitated by ammonia, filtered, and washed. The iron pre- 
 cipitate is now free from copper, and in it the iron can be 
 estimated by dissolving in nitric acid, making the solution 
 nearly neutral with ammonia, and estimating the iron by 
 the method given in the paper before referred to. The 
 filtrate from the iron precipitate is boiled till all the am- 
 monia is completely driven off, and the copper estimated 
 in the solution so obtained as already described. 
 
 2. JACQUELAIN'S AND VON HUBERT'S COLORIMETRIC ASSAYS. 
 
 Heine's method, for the reasons stated, is suitable only 
 for the estimation of small quantities of copper. Jacque- 
 lain has extended it to the examination of all cupriferous 
 substances, and this process has been further perfected 
 by Yon Hubert. According to the latter, a solution of any 
 cupriferous accurately weighed substance is prepared, 
 mixed with ammonia in excess, the ammoniacal solution 
 (assay solution) measured at a definite volume, and a small, 
 likewise measured portion of the measured solution diluted 
 with water, until its blue colour shows an equal intensity 
 with the blue colour of another solution (normal solution), 
 also cupriferous and ammoniacal, whose copper contents 
 are known once for all. Then, from the quantity of water 
 added, in order to make the two fluids equal to each other 
 in the intensity of their blue colours, the amount of copper 
 in the substance under examination can be estimated by 
 calculation. 
 
 The normal solution is prepared by dissolving -5 of a 
 
COLORIMETRIC ASSAYS. 473 
 
 gramme of chemically pure copper in dilute nitric acid, 
 adding ammonia in excess, and diluting with distilled 
 water until the whole at 12 C. amounts to one litre = 1000 
 cubic centimetres. The solution is filtered, and pre- 
 served in a flask provided with a glass stopper ground in 
 to fit it. 
 
 For the preparation of the assay fluid, with substances 
 whose percentage of copper ranges from 1-5 to the highest 
 per cent., two grammes, and with the poorer substances 
 five grammes, are brought into ammoniacal solution with 
 the precautions specified in Heine's assay. This solution, 
 with over 5 per cent, of copper, is measured at two hun- 
 dred cubic centimetres, with 2 to 5 per cent, of copper at 
 one hundred and fifty cubic centimetres, and with 2 per 
 cent, and under, at one hundred ; and also, as may be 
 required, at 90, 80, 60, 50 c.c., according to the intensity 
 of the fluid. Only with an extremely small quantity of 
 copper is the assay fluid evaporated to a smaller volume, 
 .in order to be able to conduct the colorimetric test with 
 accuracy. 
 
 The comparison of the intensity of colour of the assay 
 fluid with the normal fluid is accomplished in two different 
 ways, according as the former, when measured at a defi- 
 nite volume, is darker or lighter than the latter. This 
 can be seen if a small arbitrary portion of each is poured 
 into a glass tube of nine millimetres interior diameter, 
 twelve centimetres in length, and uniform thickness, 
 and the two tubes are held in parallel positions over a 
 piece of white paper so that they rest firmly on it, and 
 are inclined to it at an angle of about 45, and direct 
 light falls upon them. Shadow should not fall upon the 
 tubes. 
 
 (a) The Assay Fluid is Darker than the Normal 
 Solution. By means of a pipette, five cubic centimetres of 
 the normal solution are placed in a glass tube closed at the 
 bottom and not graduated, and seven millimetres in interior 
 diameter and twelve centimetres long. Since 1000 c.c. of 
 the normal solution contain -5 of a gramme of copper, five 
 cubic centimetres contain exactly '0025, and the ratio 
 
474 THE ASSAY OF COPPER. 
 
 5 : -0025 expresses once for all the known proportion of 
 copper in the normal solution. 
 
 Five cubic centimetres of the definitely measured assay 
 fluid are now also placed in a beaker and gradually 
 diluted with water till they show the same intensity of 
 colour as the normal solution. In the comparison the 
 assay fluid must be in a similar tube to that containing 
 the normal solution. With richer proportions a greater 
 accuracy is attained in this comparison if the assay fluid 
 is so far diluted that its intensity still appears as little as 
 possible darker than that of the normal solution, and then 
 water added carefully, drop by drop, till its intensity 
 is judged as little as possible lighter than that of the nor- 
 mal solution, whereupon the mean of the two volumes 
 noted is taken as the correct value. The measuring of the 
 diluted assay fluid is performed in glass tubes of nine 
 millimetres interior diameter and fifty centimetres in 
 length, which from their lower closed end to the circular 
 mark designated by 0, hold exactly five cubic centimetres, 
 and from upwards are divided into cubic centimetres 
 and tenths. 
 
 If, for example, two grammes of the assay substance 
 have been weighed out, the assay fluid measured at 200 
 cubic centimetres, and five cubic centimetres of it diluted 
 to 8-2 cubic centimetres, in order to obtain an equal inten- 
 sity of colour between the normal and assay fluid, then the 
 percentage of copper, #, follows from this according to the 
 following chain of ratios : 
 
 100 per cent. 
 
 200 c.c. assay fluid. 
 
 8-2 c.c. diluted assay = normal solution. 
 0025 gramme of copper in normal solution. 
 
 x = 8'2 per cent, of copper. 
 
 (b) The Assay Fluid is Lighter than the Normal Solution. 
 In this case five cubic centimetres of the normal solution 
 are diluted till their intensity is equal to that of the assay 
 solution that has been measured at a definite volume, and 
 for the comparison larger tubes of nine millimetres' interior 
 diameter are used. 
 
COLORIMETRIC ASSAYS. 475- 
 
 If, for example, two grammes of the assay substance 
 have been weighed out, 150 cubic centimetres of assay 
 fluid obtained from it, and to get the same intensity of 
 colour, five cubic centimetres, of the normal solution 
 diluted to 8 -4 cubic centimetres, the quantity of copper x 
 amounts, according to the following chain of ratios, to 
 2-205 per cent. : 
 
 100 per cent. 
 
 x 
 2 
 
 8-4 
 5 
 
 150 c.c. assay solution. 
 5 c.c. normal solution. 
 0025 gramme of copper. 
 
 x = 16-8 | 37-5 = 2-205 per cent. 
 
 This assay is adapted for all cupriferous substances,, 
 since nickel, cobalt, and manganese, which would influence 
 the result unfavourably, can be removed without particu- 
 lar difficulty. It is also easy to be learned by those less 
 practised in analytical operations, can be completed in a 
 few hours, and is far less expensive than the dry assay. 
 From two to one-tenth per cent, of copper can also be 
 determined by it with accuracy. 
 
 Heine, however, prefers his method when a small per- 
 centage of copper is to be estimated, since by it even one 
 loth of copper in the centner = -03 per cent., can be esti- 
 mated, and there is less liability to error. While in slag 
 assays with nine to eighteen loth of copper in the centner,, 
 by Heine's method, errors of half a loth are not to be 
 avoided, variations of more than one loth occur by Von 
 Hubert's process. The latter works with a too deeply 
 coloured normal fluid, corresponding to a solution of over 
 fourteen loth of copper to one ounce of water, while Heine 
 does not exceed four loth. The process is surer if the fluids 
 are diluted and thicker layers of them compared, and the 
 hue thus made artificially deeper, than if small quantities of 
 stronger fluids are compared and the hue made artificially 
 lighter by comparing them in thinner layers, or especially 
 in tubes, where the light is dispersed and shade produced. 
 The comparison in oblong glasses is therefore to be pre- 
 ferred to that in tubes. 
 
476 THE ASSAY OF COPPER. 
 
 By a comparison of Yon Hubert's assay with tnat of 
 the Oberhartz, it appears that, as Yon Hubert's experi- 
 ments themselves have shown, both give equally accurate 
 results for substances not too poor in copper (i.e. contain- 
 ing not less than 5 per cent.) The Oberhartz assay 
 allows a direct estimation of the copper, requires less ap- 
 paratus, is also very simple, and can be completed in a 
 shorter time. Since different individuals are differently 
 susceptible to colours, and the blue colour of the copper 
 ammoriio-oxide, in consequence of causes yet unknown, 
 sometimes inclines more or less to greenish, and thereby 
 renders observation difficult, therefore, for the sake of 
 greater certainty, though not of greater accuracy, those 
 assays by which an estimation of the copper is possible 
 by weight should in general be preferred to the colori- 
 metric methods, and this is the case with the Oberhartz 
 assay down to two per cent. With smaller percentages the 
 colorimetric assay must be called to our aid. It is not 
 yet settled that with higher percentages of copper the 
 principle of colorimetry is a correct one ; that is, that the 
 intensity of the colour is directly proportional to the quan- 
 tity of the colouring agent. 
 
 Since ammoniacal solutions poor in copper often show 
 a dash of green colour, Yon Hubert prepares a normal 
 solution for such by dissolving one decigramme of copper 
 and diluting to one litre of fluid. 
 
 b. Volumetric Copper Assays. 
 FLECK'S MODIFICATION OF MOHR'S METHOD.* 
 
 The proposal to take the action of solution of potas- 
 .sium cyanide on ammoniacal solution of copper, as the 
 foundation of a method for estimating copper, is due to 
 Carl Mohr.f 
 
 * This process is given by Fresenius, condensed from the ' Polytechn. 
 Oentralbl.' 1856, 1313. 
 
 t Annal. d. Chem. u. Pharm. 94, 198 ; Fr. Mohr's Lehrbuch der Titrier- 
 rnethode, 2, 91. 
 
VOLUMETRIC ASSAYS. 477 
 
 In carrying out this estimation according to the direc- 
 tions of Mr. Parkes, a solution of potassium cyanide is slowly 
 added to a blue ammoniacal solution of copper, when the 
 latter gradually loses its colour, and finally becomes quite 
 colourless ; upon this chemical reaction the estimation of 
 copper by cyanide of potassium depends. By ascertaining 
 by direct experiment the amount of potassium cyanide 
 solution required to discharge the colour in an ammoniacal 
 solution containing a given weight of copper, it is easy 
 by a comparative experiment to estimate the amount of 
 copper in a given weight of ore. 
 
 For the preparation of the standard solution 2,000 
 grains of fused potassium cyanide are to be dissolved in 
 two quarts of water, to produce a solution of which 1,000" 
 grains measure will be equal to about ten grains of metallic 
 copper. The solution should be preserved in green glass 
 stoppered bottles, and kept as much as possible away from 
 the light : it is liable to a slow decomposition, which will 
 necessitate the standard being checked at intervals of one 
 or two weeks. In order to standardise the solution, a 
 burette, holding 1,000 grains measure, is filled to the zero 
 mark, and a piece of pure electrotype copper, previously 
 cleaned by means of dilute nitric acid, washed and dried, 
 is accurately weighed. About eight grains may be con- 
 veniently taken ; this is dissolved in a pint flask by dilute 
 nitric acid, and, after the energy of the first action has sub- 
 sided, the solution is warmed and ultimately boiled to expel 
 all the nitrous acid fumes. It is diluted with cold water 
 to the bulk of nearly half a pint, treated with ammonia in 
 excess, and to the deep blue solution the cyanide is added 
 from the burette until the colour is so nearly discharged 
 that a faint lilac tint only remains. This will generally 
 become quite bleached on standing at rest for a short time, 
 so that the cyanide must not be added too hastily towards 
 the end of the operation. It will be advisable to control 
 the standard by a second experiment upon another weighed 
 portion of copper, and to stop short of bleaching entirely 
 the faint lilac tint of the solution. A piece of white paper 
 folded and placed under and behind the flask during the 
 
478 THE ASSAY OF COPPER. 
 
 decolorisation, will aid in recognising the proper tint of 
 the solution. 
 
 In applying this process to the examination of copper 
 ores, a known weight of the finely powdered sample is 
 introduced into a beaker provided with a glass cover, and 
 moistened with strong sulphuric acid ; strong nitric acid is 
 then added, and the whole digested on a sand-bath until 
 nitrous fumes are no longer given off. Should a small 
 quantity of sulphur be separated in the treatment of py- 
 ritic ores, the small globules may be taken out, burnt, and 
 the residual copper dissolved in a few drops of nitric acid 
 and mixed with the remainder. Water is now to be 
 added and left in contact for a short time to extract all 
 the metallic salt from the insoluble residue, which need not 
 be filtered off ; and so, likewise, when ammonia is added 
 in the next place, any ferric oxide which may thus be 
 precipitated is left in the solution, for it is apt to contain a 
 small proportion of copper when first thrown down ; but 
 this is entirely removed by the potassium cyanide later in 
 the experiment. 
 
 When the ore contains much iron it is considered 
 desirable to remove the hydrated peroxide by filtration, in 
 order to be enabled to observe with greater precision the 
 last effects of the potassium cyanide ; and in the event of 
 requiring to know the amount of iron present in the ore, 
 the precipitated ferric oxide on the filter is redissolved in 
 dilute sulphuric acid, reduced to the state of ferrous oxide 
 by metallic zinc, and then tested in the usual way with a 
 standard solution of potassium bichromate. 
 
 The metals which interfere with this mode of valuing 
 copper ores, are silver, nickel, cobalt, and zinc. The first 
 may readily be separated by adding a few drops of hydro- 
 chloric acid to the original solution : the other metals may 
 be excluded by following one of the methods pointed out 
 by the author for that purpose. 
 
 Fleck proposes the following modification in this pro- 
 cess : 
 
 Instead of caustic ammonia, use a solution of ammo- 
 nium sesquicarbonate (1 in 10), warm the mixture to 
 
VOLUMETEIC ASSAYS. 479 
 
 about 60, and, in order to render the end reaction plainer, 
 add 2 drops of solution of potassium ferrocyanide (1 in 
 20) : the blue colour of the solution is not altered by this 
 addition, nor is its clearness affected. The value of the 
 potassium cyanide solution is first estimated by means of 
 copper solution of known strength, and it is then employed 
 on the copper solution to be examined. On dropping the 
 potassium cyanide into the blue solution warmed to 60, 
 the odour of cyanogen is plainly perceptible, and the 
 colour gradually disappears. As soon as the ammoniacal 
 double salt of copper is destroyed, the solution becomes 
 red from the formation of copper ferrocyanide, without 
 any precipitate appearing, and with the addition of a 
 final drop of potassium cyanide this red colour in its turn 
 vanishes, so that the fluid now appears quite colourless. 
 
 The method thus modified yields, it is true, better, but 
 still only approximate, results.* Where such are good 
 enough, the method is certainly convenient. 
 
 E. 0. BROWN'S METHOD BY SODIUM HYPOSULPHITE. 
 
 The process described by Mr. E. 0. Brown is particu- 
 larly applicable to the estimation of copper in gun- 
 metal, brass, and other alloys which contain no large 
 amounts of iron and lead. It is founded on the reactions 
 between salts of copper and the neutral iodides, and on 
 the conversion of the liberated iodine into hydriodic and 
 tetrathionic acids by a standard solution of sodium hypo- 
 sulphite. 
 
 The completion of the second reaction is manifested 
 by the bleaching effect produced upon the blue iodide of 
 starch by the addition of the hyposulphite. A convenient 
 strength of solution for this purpose may be made by dis- 
 solving 1,300 grains of the crystallised salt in two quarts 
 of water. The potassium iodide must be free from iodate ; 
 and a clear solution of starch employed. 
 
 * In six experiments, in which he had purposely added different quantities 
 of carbonate of ammonia, Fleck used for 100 c.c. copper solution, in the 
 minimum 15-2, in the maximum 15-75, in the mean 15'46 c.c potassium 
 cyanide solution. 
 
480 THE ASSAY OF COPPER. 
 
 From eight to ten grains of the copper or alloy are 
 dissolved in dilute nitric acid, and the red nitrous fumes 
 expelled by boiling. The copper nitrate is converted into 
 acetate by adding sodium carbonate until a portion of 
 copper remains precipitated, and then re-dissolving in 
 acetic acid. The solution is diluted with water, and about 
 60 grains of potassium iodide in the form of crystals 
 dropped into the flask, and allowed to dissolve. The 
 standard solution of sodium hyposulphite is now poured 
 in from a burette, until the greater part of the dark- 
 coloured free iodine disappears. A little of the starch 
 solution is now added to make its presence more apparent, 
 and the addition of the hyposulphite continued until the 
 bleaching is completed, when the pale yellow colour of the 
 copper subiodide will alone be visible. The amount of 
 copper in the ore or alloy is calculated from the number 
 of divisions indicated upon the burette. 
 
 Copper ores containing much iron (which interferes by 
 reason of the dark red colour of the acetate) may be dis- 
 solved in nitric acid, and treated with sulphuretted hydro- 
 gen to precipitate the copper, the sulphide being collected 
 on a filter, washed, and re-dissolved in nitric acid to pro- 
 duce a solution suitable for testing by this process. Or 
 the hyposulphite may itself be employed to furnish a pre- 
 cipitate of copper disulphide. 
 
 c. Electrolytic Copper Assay. 
 
 ESTIMATION OF COPPER IN THE MANSFELD ORES BY 
 DR. STEINBECK'S PROCESS. 
 
 This method embraces three distinct operations, viz. : 
 1. The extraction of the copper from the ore; 2. The 
 separation ; 3. The quantitative estimation of that metal. 
 
 1. The Extraction of the Copper from the Ore. A proof 
 centner, equal to 5 grammes of pulverised ore, is put into 
 a flask, and there is poured over it a quantity of from 40 
 to 50 c.c. of crude hydrochloric acid, of a specific gravity 
 of 1-16, whereby all carbonates are converted into chlorides 
 
STEINBECK S PROCESS. 481 
 
 while carbonic acid is expelled. After a while there is added 
 to the fluid in the flask 6 c.c. of a normal nitric acid, pre- 
 pared by mixing equal bulks of water and pure nitric acid 
 of 1*2 sp. gr. As regards certain ores, however, specially 
 met with in the district of Mansfeld, some, having a very 
 high percentage of sulphur and bitumen, have to be roasted 
 previously to being subjected to this process ; and others, 
 again, require only 1 c.c. of nitric acid instead of 6. The 
 flask containing the assay is digested on a well-arranged 
 sand-bath for half an hour, and the contents only boiled 
 for about fifteen minutes, after which the whole of the 
 copper and other metals occurring in the ore are in 
 solution as chlorides. The blackish residue, consisting of 
 sand and schist, has been proved by numerous experi- 
 ments to be either entirely free from copper, or at the 
 most only 0-01 to O03 per cent, has been left undissolved. 
 
 The extraction of the copper from the ore, according 
 to this method, is complete even in the case of the best 
 quality of ore, which contains about 14 per cent, of metal ; 
 while, at the same time, the very essential condition for 
 the proper and complete separation of the metal, viz. the 
 entire absence of nitric acid, or any of the lower oxides 
 of nitrogen, is fully complied with. 
 
 2. Separation of the Copper. The solution of metallic 
 and earthy chlorides, and some free hydrochloric acid, 
 obtained as just described, is separated by filtration from 
 the insoluble residue, and the fluid run into a covered 
 beaker of about 400 c.c. capacity ; in this beaker has 
 been previously placed a rod of metallic zinc, weighing 
 about 50 grammes, and fastened to a piece of stout pla- 
 tinum foil. The zinc to be used for this purpose should 
 be as much as possible free from lead, and at any rate 
 not contain more than from 0-1 to 0*3 per cent, of the 
 latter metal. The precipitation of the copper in the me- 
 tallic state sets in already during the filtration of the warm 
 and concentrated fluid, and is owing chiefly to the com- 
 plete absence of nitric acid completely finished in from 
 half to three-quarters of an hour after the beginning of 
 the filtration. If the fluid be tested with sulphuretted 
 
 I I 
 
482 THE ASSAY OF COPPEE. 
 
 hydrogen, no trace even of copper will be detected ; the 
 spongy metal partly covers the platinum foil, partly floats 
 about in the liquid, and, in case either the ore itself or 
 the zinc applied in the experiment contained lead, small 
 quantities of that metal will accompany the precipitated 
 copper. After the excess of zinc (for an excess must be 
 always employed) has been removed, the spongy metal is 
 repeatedly and carefully washed by decantation with fresh 
 water, which need not be distilled, and care is taken to 
 collect together every particle of the spongy mass. 
 
 3. Quantitative Estimation of the Precipitated Copper. 
 To the spongy metallic mass in the beaker, wherein the 
 platinum foil is left, since some of the metal adheres to it, 
 8 c.c. of the normal nitric acid are added, and the copper 
 is dissolved, by the aid of moderate heat, as copper 
 nitrate : in the event of any small quantity of lead being 
 present, it will of course be contaminated with lead 
 nitrate. 
 
 When copper ores are dealt with which contain above 
 6 per cent, of copper, which may be sufficiently judged 
 from the larger bulk of the spongy mass of precipitated 
 metal, 16 c.c. of nitric acid, instead of 8, are employed 
 for dissolving the metal. The solution thus obtained 
 is left to cool, and next, immediately before titration 
 with potassium cyanide, mixed with 10 c.c. of normal 
 solution of ammonia, prepared by diluting 1 volume of 
 liquid ammonia, sp. gr. 0*93, with 2 volumes of distilled 
 water. 
 
 In the case of ores which yield over 6 per cent, of 
 copper, and when a double quantity of normal nitric acid 
 has consequently been used, the solution of copper in 
 nitric acid is diluted with water, and made to occupy a 
 bulk of 100 c.c. ; this bulk is then divided exactly into 
 two portions of 50 c.c. each, and each of these separately 
 mixed with 10 c.c. of the liquid ammonia solution just 
 alluded to, and the copper therein volumetrically esti- 
 mated. The deep blue-coloured solution of oxide of 
 copper in ammonia only contains, besides ammonium 
 nitrate, any lead which might have been dissolved, having 
 
STEINBECK'S PROCESS. 483 
 
 been precipitated as hydrated lead oxide, which does not 
 interfere with the titration with potassium cyanide. The 
 solution of the last-named salt is so arranged that 1 c.c. 
 thereof exactly indicates 0-005 grm. of copper. Since, 
 for every assay, 5 grins, of ore have been taken, 1 c.c. of 
 the titration fluid is, according to the following, proportion 
 5 : O'OOo :: 100 : O'l equal to O'l per cent, of copper ; 
 it hence follows that, by multiplying the number of the 
 c.c. of potassium cyanide solution used to make the blue 
 colour of the copper solution disappear, by O'l, the per- 
 centage of copper contained in the ore is immediately 
 indicated. 
 
 As may be imagined, at the laboratory of the mine- 
 owners at Eisleben, such a large number of assays are 
 daily executed that, in this case, there can be no reason to 
 fear a deterioration of the cyanide solution, of which large 
 quantities are used and often fresh made ; but for security's 
 sake the solutions are purposely tested for control at least 
 once every week. According to the described plan, six 
 assays can be made within 4 hours ; and during a work- 
 ing day of from 7^ to 8 hours, twenty assays have been 
 often quite satisfactorily made by the umpires, as well as 
 by the workmen at Eisleben. 
 
 Special Observations on this Method. Dr. Steinbeck 
 considered it necessary to test this method specially, in 
 order to see what influence is exercised thereupon by (1) 
 ammonium nitrate, (2) caustic ammonia, (3) the presence of 
 lead oxide. The copper used to perform the experiments 
 for this purpose was pure metal, obtained by galvano- 
 plastic action, and was ignited to destroy any organic 
 matter which might accidentally adhere to it, and, next, 
 cleaned by placing it in dilute nitric acid. Five grammes 
 of this metal were placed in a litre flask, and dissolved in 
 266 '6 c.c. of normal nitric acid, the flask and contents 
 gently heated, and, after cooling, the contents diluted with 
 water, and thus brought to a bulk of 1,000 c.c. exactly. 
 Thirty c.c. of this solution were always applied to test and 
 titrate one and the same solution of potassium cyanide 
 under all circumstances. When 5 grammes of ore, con- 
 
 i i 2 
 
484 THE ASSAY OF COPPER. 
 
 taining on an average 3 per cent, of copper, are taken for 
 assay, that quantity of copper is exactly equal to 0-15 
 gramme of the chemically pure copper. The quantity of 
 normal nitric acid taken to dissolve 5 grammes of pure 
 copper (266 -6 c.c.) was purposely taken, so as to corre- 
 spond with the quantity of 8 c.c. of normal nitric acid 
 which is applied in the assay of the copper obtained from 
 the ore, and this quantity of acid is exactly met with in 
 30 c.c. of the solution of pure copper. 
 
 As regards No. 1 and No. 2 (see above), the influence 
 of double quantities of ammonium nitrate and free caustic 
 ammonia (the quantity of copper remaining the same), and 
 the action of dilute solution of potassium cyanide there- 
 upon, will become elucidated by the following facts : 
 
 a. Thirty c.c. of the normal solution of copper, con- 
 taining exactly 0-15 gramme of copper, were rendered 
 alkaline with 10 c.c. of normal ammonia, and are found 
 to require, for entire decoloration, 29-8 c.c. of potassium 
 cyanide solution. A second experiment, again with 30 c.c, 
 of normal copper solution, and otherwise under identically 
 the same conditions, required 29-9 c.c. of cyanide solution. 
 The average of the two experiments is 29*85 c.c. 
 
 b. When to 30 c.c. of the normal copper solution 8 c.c. 
 of normal nitric acid are first added, and then 20 c.c. of 
 normal ammonia solution, instead of only 8, whereby the 
 quantity of free ammonia and of ammonium nitrate is 
 made double what it was in the case of the experiments 
 spoken of under a, there is required of the same cyanide 
 solution 30-3 c.c. to produce decoloration. A repetition 
 of the experiment, under exactly the same conditions, gave 
 30-4 c.c. of the cyanide solution employed ; the average 
 of both experiments is, therefore, 30-35 c.c. 
 
 The difference between 30-35 and 29-85 is equal to 
 0-5 c.c., and that figure is therefore the coefficient of the 
 influence of double quantities ; and supposing this to 
 happen with the ores in question, it would only be equiva- 
 lent to 0-05 per cent, of metallic copper. It is hence 
 clear that slight variations of from 0-1 to 0'5 c.c. in 
 the measuring out of 8 c.c. of normal nitric acid, used 
 
STEINBECK S PEOCESS. 485 
 
 to dissolve the spongy copper, and of 10 c.c. of normal 
 ammonia, in order to render the nitric acid copper solu- 
 tion alkaline, are of no consequence whatever for the 
 technical results to be deduced from the assay. It should, 
 moreover, be borne in mind that the quantities of free 
 .ammonia and of ammonium nitrate in the actual assay of 
 ores, for which always a quantity of 5 grammes of ore is 
 taken, vary according to the richness or poverty of the 
 ores in copper ; and the quotation of the following results 
 of experiments proves that the influence of these sub- 
 stances is only very slightly felt in the accuracy of the 
 results : 
 
 Eight c.c. of the normal nitric acid have been found 
 to contain, by a series of experiments, 1*353 gramme 
 of anhydrous nitric acid ; and this quantity of acid is 
 -exactly neutralised by 7*7 c.c. of normal ammonia solu- 
 tion, which contains 0-6515 gramme of ammonium oxide ; 
 and 10 c.c. of the said normal solution contain 0-846 
 gramme of ammonium oxide. 
 
 One gramme of metallic copper requires, for complete 
 oxidation, 0-2523 gramme of oxygen, and this quantity of 
 -oxygen is given off by 0*5676 gramme of anhydrous nitric 
 acid ; while, at the same time, binoxide of nitrogen is 
 disengaged. From these data can be calculated (1) the 
 quantity of nitric acid which becomes decomposed when 
 variable quantities of metallic copper are dissolved 
 therein; (2) what quantity of nitric acid is left to form 
 neutral nitrate of ammonium ; and (3) what quantity of 
 free ammonia will be left after a portion of that alkali has 
 been combined with, and therefore neutralised by, copper 
 oxide ; and any remaining free nitric acid. 
 
 It is found that the quantitative variations between 
 ores containing 1 per cent, or 6 per cent, of metal vary 
 very little from the normal quantities exhibited by ores 
 containing 3 per cent, of metal. The relation is as 1 : 2 ; 
 -and, for technical purposes, this has been proved not to 
 be a disturbing quantity. 
 
 When, however, larger quantities of ammoniacal salts 
 are present in the fluid to be assayed for copper by means 
 
486 THE ASSAY OF COPPER. 
 
 of a titrated solution of potassium cyanide, and especially 
 when ammonium carbonate, sulphate, and, worse still, 
 chloride are simultaneously present, these salts exert a 
 very disturbing influence. 
 
 The presence of lead oxide in the copper solution to 
 be assayed has the effect of producing, on the addition of 
 10 c.c. of normal ammonia, a milkiness along with the 
 blue tint ; but the presence of this oxide does not at all 
 interfere with the estimation of the copper by means of 
 the cyanide, provided the lead be not in great excess ; and 
 a slight milkiness of the solution even promotes the visi- 
 bility of the approaching end of the operation. 
 
 Dr. Steinbeck, however, purposely made some experi- 
 ments to test this point, and his results show that neither 
 50 nor 100 per cent, of addition of lead exerts any per- 
 ceptible influence upon the estimation of copper, from its 
 ores or otherwise, by means of potassium cyanide. A 
 small quantity of accidentally occurring lead will not, 
 therefore,- affect the results, and this the less so as, gener- 
 ally, no ores of both metals occur together wherein both 
 are met in sufficient quantity to make it worth while 
 working the ore for both metals at the same time. 
 
 Since it is well known that the presence of zinc very 
 perceptibly influences the action of a solution of potassium 
 cyanide, when applied to the volumetrical estimation of 
 copper, Dr. Steinbeck considered it necessary to institute 
 some experiments in order precisely to ascertain with 
 what quantity of zinc present along with copper this in- 
 fluence commences to become perceptible. The solution 
 of zinc applied was made by dissolving the metal in the 
 smallest possible quantity of nitric acid ; and 1 c.c. of that 
 solution contained 0*001 gramme of zinc. The results of 
 the experiments show that the presence of zinc does not 
 interfere with the visibility of the end of the reaction, viz. 
 the decoloration of the copper solution. They also prove 
 that a small quantity of zinc, less than 5 per cent, of the 
 quantity of copper present, or 0*0075 gramme by weight 
 of zinc, does not at all affect the action of the solution 
 of potassium cyanide ; but when the quantity of zinc 
 
- LUCKOW'S PROCESS. 487 
 
 increases, a very perceptible effect is seen upon the 
 solution of cyanide : it is therefore necessary to bestow 
 due care while washing the spongy copper after it has 
 been precipitated by means of zinc from its solution. 
 
 Since it has been ascertained that the action of the 
 solution of potassium cyanide in researches of thi's kind 
 is also affected by an increased temperature of the solu- 
 tion of copper which is to be titrated, it is strictly neces- 
 sary never to operate with warm ammoniacal solutions of 
 copper, but to suffer the same to cool down to the ordinary 
 temperature of the air of the laboratory. 
 
 While 30 c.c. of copper solution, containing 0*15 
 gramme of copper, and 10 c.c. of normal ammonia solu- 
 tion, required at the ordinary temperature 30 c.c. of 
 cyanide solution, the same quantities required, at between 
 40 and 45 C., 28*8 c.c. of solution of cyanide ; and at. 45 
 C., 28'9 c.c. of the same solution, thus proving the injurious 
 effect of warm solutions. 
 
 ESTIMATION OF COFFEE IN THE MANSFELD ORES BY 
 M. C. LUCKOW'S PROCESS. 
 
 This gentleman has applied to the quantitative estima- 
 tion of copper a new method, based upon the. precipitation 
 of the metal in the metallic state from solutions containing 
 either free sulphuric or nitric acid, by means of a galvanic 
 current. 
 
 It is a great advantage of this method, that, while the 
 copper is precipitated, it is simultaneously separated from 
 metals witji which it is often found alloyed ; some of these, 
 such as tin and antimony, are separated by treatment with 
 nitric acid in an insoluble form ; while others, like silver, 
 can easily be removed in the form of chloride. It is,, at 
 the same time, another advantage that the state in which 
 the copper is obtained admits of its being accurately 
 weighed and estimated, while a great number of opera- 
 tions, which require much time and various apparatus, are 
 at the same time got rid of. 
 
 Although M. Luckow had previously discovered a 
 
488 THE ASSAY OF COPPER. 
 
 method of electro-metallic analysis from fluids containing 
 free sulphuric acid, his researches on the same subject, 
 in the case of free nitric acid, belong to a recent period. 
 These researches brought very unexpectedly to light the 
 curious fact that even a weak galvanic current had the 
 power of completely precipitating copper in a pure metal- 
 lic state from nitric solutions, provided they did not con- 
 tain more than O'l gramme of anhydrous nitric acid to 
 the c.c. (nitric acid of 1'2 sp. gr. contains 0-32 gramme 
 of anhydrous nitric acid to the c.c.) ; while it was found 
 that the action was, at the same time, more regular, and 
 less dependent upon the power of the current than when 
 free sulphuric acid was present. The following more com- 
 monly occurring metals are not precipitated by galvanic 
 action from acid solutions : Zinc, iron, nickel, cobalt, 
 chromium, the metals of the earths and alkalies. The fol- 
 lowing are precipitated (a) in the shape of peroxides, at the 
 positive electrode : completely, lead and manganese ; in- 
 completely, silver. When the solution contains traces of 
 manganese, it becomes, in consequence of the formation of 
 a salt of manganese peroxide, or of permanganic acid, deeply 
 violet-coloured. This very sensitive reaction for man- 
 ganese also takes place when small quantities of chlorine 
 are present. The presence in the fluid of oxalic and 
 tartaric acids, and other readily oxidisable organic sub- 
 stances, and such protoxides as are readily peroxidised 
 for instance, iron protoxide retards the formation of 
 peroxides, as well as the occurrence of the reaction of 
 manganese. 
 
 (b) Mercury, silver, copper, and bismuth are precipi- 
 tated at the negative electrode in a metallic state. When 
 mercury is present in the solution simultaneously with 
 copper, the former metal is separated before the latter, in 
 the fluid metallic state. As soon, however, as the pre- 
 cipitation of copper commences, an amalgam of the two 
 metals is formed, when mercury is also present. Silver is 
 precipitated almost simultaneously with copper ; bismuth 
 only begins to be precipitated after the greater portion of 
 the copper has been separated. A complete separation of 
 
LUCKOWS PROCESS. 481) 
 
 silver only ensues when some such substance as tartaric or 
 other similar acid is simultaneously present in the solution. 
 The separation of the three last-named metals by means 
 of galvanic action is, therefore, unsuccessful ; but, fortu- 
 nately, there are many other means to accomplish this end 
 completely. 
 
 (c) Metallic arsenic is only precipitated slowly, and 
 long after the complete separation of copper, if arsenic 
 acid happen to be present. The same remark applies to 
 antimony, since it is well known that small quantities of 
 antimonic acid are soluble in nitric acid. 
 
 The operations, according to Luckow's plan, are 
 1. Eoasting the ore; 2. Solution of the roasted product; 
 3. Precipitation of the copper ; 4. Weighing the copper. 
 
 1. Eoasting the Ore. Care should be taken to obtain 
 a finely ground average sample of the ore. Then weigh 
 off in small porcelain capsules, previously counterpoised, 
 quantities of from 1 to 3 grammes ; these quantities are then 
 placed on the inverted lid of an iron crucible, on the inner 
 surface of which the powdered ore is heated over the flame 
 of a Bunsen gas burner. The powder may be carefully 
 stirred up with a platinum wire, to promote the access of 
 air during the roasting ; the ignition of bituminous matter 
 and sulphur will be ended in about seven minutes. Ores 
 which do not contain bitumen at all need not be roasted. 
 
 It has been already stated that in the case of poor 
 copper ores (and those of the Mansfeld district are gener- 
 ally so), the quantities to be weighed off for assay should 
 not vary according to a presumed percentage of copper. 
 Two grammes are therefore taken, and, instead of roasting 
 the ore on the lid of an iron crucible, small porcelain 
 crucibles are used for that purpose. 
 
 2. Solution of the Roasted Product. The iron lid is 
 suffered to cool, the roasted powder placed on a piece of 
 glazed paper, and any powder adhering to the lid is re- 
 moved by means of a camel's- hair brush on to the paper. 
 The powder is next transferred to small beaker glasses, 
 and about 2 or 3 c.c. of nitric acid, of 1-2 specific gravity, 
 are added, along with about 10 to 15 drops of concentrated 
 
490 THE ASSAY OF COPPER. 
 
 sulphuric acid. The beakers are then placed on a sand- 
 bath and moderately heated, at first ; but when the con- 
 tents have become nearly dry, the heat is increased, so as 
 to evaporate and expel all sulphuric acid. The beakers 
 should be covered with perforated watch-glasses. This 
 operation requires from about three-quarters of an hour 
 to one hour. The addition of sulphuric acid is made in 
 order to increase the oxidising action of the nitric acid, 
 and also to convert any lime which may happen to be 
 present in the ore into a difficultly soluble salt. It is very 
 useful, also, to add from 10 to 20 drops of hydrochtoric 
 acid to the mixture of the two acids just alluded to, since 
 the rapidity of the evaporation is thereby increased, and 
 the occasional spirting about of the fiuid is lessened. 
 
 The process just described may be modified, first, by 
 the use of porcelain capsules, the contents of which are 
 easily transferred to beakers with flat bottoms, and not 
 higher than about 2 inches altogether. It is better, also, 
 to use sulphuric acid, prepared with equal bulks of con- 
 FIG. 105. centrated acid and water, and to measure 
 off 4 c.c. for each assay ; while for each 
 assay, moreover, 6 c.c. of nitric acid and 
 about 25 drops of hydrochloric acid are 
 taken. Instead of covering the beaker 
 with a perforated watch-glass, the upper 
 part of a funnel is used, as represented 
 
 in fig. 105 ; with this arrangement the sulphuric acid 
 evaporates far more readily, and loss by spirting is pre- 
 vented. The beaker is heated on a well-arranged sand- 
 bath. 
 
 3. Precipitation of the Copper. As soon as the beaker 
 after removal from the sand-bath has become quite cool, 
 the funnel which has been used as a cover is washed on 
 both sides, inner and outer, with nitric acid of 1*2 sp. gr., 
 diluted with six times its bulk of pure water ; the sides of 
 the beaker are next likewise washed, and it is then filled 
 to about half its height with the same acid. A few drops 
 of a concentrated solution of tartaric acid are added (this 
 acid is best kept in solution in open vessels, only slightly 
 
LUCKOWS PROCESS. 
 
 491 
 
 covered with a piece of paper) ; this having been done, 
 the wire spiral, represented in fig. 106, is carefully placed 
 in the beaker. This spiral consists of a piece of platinum 
 wire, about l-12th of an inch thick, and 7-J- inches long, 
 two-thirds of its length being so wound that the straight 
 end of the wire projects as if it were the axis of the centre 
 of the spiral. The convolutions of the spiral are so large 
 that they touch the sides of the beaker, while the straight 
 portion just touches the centre of the bottom of the vessel. 
 When the heating has been carefully attended to, 
 the acid fluid added to the contents of the beaker, after 
 evaporation to dryness, will generally be quite clear ; if it 
 
 FIG. 106. 
 
 FIG. 107. 
 
 FIG. 108. 
 
 happens to be turbid, 1 or 2 c.c. of a concentrated solution 
 of barium nitrate may be added, and the thorough mixing 
 of this saline solution with the acid contents of the beaker 
 promoted by gently moving up and down the platinum 
 spiral just alluded to, and allowing the fluid to rest for -a 
 few minutes after. The copper present in the mass left 
 at the bottom of the beaker gradually dissolves, and it is 
 not actually requisite to wait before applying the galvanic 
 current until it is all dissolved. 
 
 The next point is to place in the beaker the platinum 
 
492 THE ASSAY OF COPPER. 
 
 foil, represented at fig. 107, of which the dimensions are 
 length, 2^ inches ; width, 1J inch. The lower end of this 
 platinum foil should be kept about 1-1 6th of an inch apart 
 from the convolutions of the spiral. When the beaker is 
 only half filled with liquid, the platinum foil is immersed 
 in the same for more than three-fourths of its height. The 
 wire fastened to this foil is fixed, by means of a screw a, 
 to the arm a b of the stand, represented in fig. 108 ; the 
 other screw, b, serves to fasten a copper wire proceeding 
 from the zinc end of the galvanic battery. When the 
 small screw clamp c (fig. 108) has been fastened to the 
 platinum wire placed in the beaker, another wire is 
 fastened in the top opening of the clamp, and this wire 
 connected with the copper end of the battery, and the 
 galvanic circuit thus closed. In a few moments after this 
 has been done, the platinum foil, bent in the shape of the 
 cylinder and placed inside the beaker, as before described, 
 will be observed to become covered with a coating of 
 metallic copper, while from the platinum wire spiral 
 bubbles of gas escape, which facilitate, to some extent, the 
 solution of the copper oxide in the dilute acid. 
 
 In order to ascertain whether the whole quantity of 
 the copper has been precipitated, some more dilute nitric 
 acid is added to the fluid in the beaker glass. If, in ten 
 minutes after this, no more metallic copper is separated on 
 the clean portions of the platinum foil, the operation is 
 finished. 
 
 It must be here observed that continued practice has 
 proved that the addition of a concentrated solution of 
 barium nitrate acts injuriously on the process, as the 
 metallic copper, which becomes separated, gets mixed 
 with some insoluble barium sulphate, which increases the 
 weight of the substance to be weighed. 
 
 The time occupied by the complete precipitation of the 
 metal varies according to the force of the galvanic current. 
 It takes from three to even eight hours. In order to make 
 this point certain, all test assays are left, for eight hours 
 consecutively, to the action of the galvanic current, expe- 
 rience having proved that, after that lapse of time, even 
 
LUCKOWS PROCESS. 493 
 
 with a weak current, the precipitation was so complete 
 that all chemical reagents for detecting the presence of 
 copper failed to discover the most minute trace of that 
 metal. 
 
 4. Weighing the Copper. The platinum cylinder to 
 which the copper adheres, and the platinum wire spiral, 
 are disconnected from the galvanic apparatus, the plati- 
 num cylinder carefully removed from the beaker and 
 immediately plunged into a beaker filled with fresh cold 
 water, and rinsed therein ; next washed with alcohol, by 
 means of a washing bottle, and then dried in a drying ap- 
 paratus, and weighed after cooling. Since the platinum 
 cylinder has been- very accurately weighed before the 
 experiment, its increase in weight will, of course, be that 
 of the copper obtained. 
 
 The process here described has been somewhat modi- 
 fied and greatly improved upon at Eisleben, where it is in 
 constant use, by the employment of a series of galvanic 
 elements. It is, in the first place, found better not to 
 disconnect the galvanic current while the copper is yet in 
 contact with acid, so that, instead thereof, the acid fluid in 
 the beaker is replaced by turning in a stream of water,, 
 and suffering the same to run over the sides of the beaker, 
 and to be received into a proper vessel to hold it. In 
 this manner all the acid is displaced, without risk of 
 any very small quantity of copper becoming acted upon 
 by the acid during the brief period elapsing between the 
 disconnecting of the galvanic current and the removal 
 from the beaker of the platinum cylinder and spiral wire. 
 These parts, on being removed, are carefully washed, first 
 with boiling water, next with alcohol, and then dried at a 
 temperature of about the boiling-point of water. The 
 cylinder is then weighed, the copper coating is removed 
 by means of nitric acid ; the platinum is next washed in 
 water, dried, and again weighed. 
 
 There are in use at Eisleben nine galvanic batteries 
 (lead and zinc elements) ; these yield eighteen assays ready 
 for weighing in twenty-four hours ; and it would not be 
 difficult for the person there employed to work with twelve- 
 
494 THE ASSAY OP COPPER. 
 
 batteries each of three elements. In place of the Mei- 
 dinger elements, which do not remain constant for months, 
 a thermo-electric apparatus has been introduced, with the 
 best results. The results obtained are highly satisfactory. 
 The following observations may be made in reference to 
 this method : 
 
 (a) The quantity of ore taken for trial is 2 grammes ; 
 this is found sufficient, while it consumes less acid. 
 
 (b) The evaporation of the acid is carried on to com- 
 plete dryness on the sand-bath. Spirting of the fluid is 
 easily prevented. 
 
 When the copper has been precipitated properly it 
 will show its peculiar colour on the surface, and the good 
 success of the operation may also be judged from the fact 
 that no saline matter adheres to the platinum ; the com- 
 plete absence of this saline matter has been found to 
 be evidence of perfect removal of the copper from the 
 fluid. 
 
 The process just described is especially applicable for 
 rather poor ores, such as do not contain above 7 or 8 per 
 cent, of copper. Each assay, from beginning to end, takes 
 ten hours for complete analysis ; but it is evident that the 
 greater portion of this period does not give active employ- 
 ment to the assayer. The expense of working this process, 
 after the apparatus has been once purchased, is very small. 
 The process may also be applied to analyse richer ores, 
 and also alloys of copper, with some slight modifications 
 which will readily suggest themselves. 
 
 Assay of Copper Pyrites. The following method of 
 treating copper pyrites has been found more advantageous 
 than the ordinary process of oxidising the mineral with 
 aqua regia, and subsequently evaporating the solution re- 
 peatedly with hydrochloric acid, or with sulphuric acid, to 
 expel the last traces of nitric acid. It is thus described 
 by Mr. F. P. Pearson in the ' Chemical News ' : 
 
 Place a weighed quantity of the powdered mineral, to- 
 gether with some potassium chlorate, in a porcelain dish. 
 (Five grammes of a variety of a pyrites containing about 
 18 per cent, of copper was found to be enough for one 
 
ASSAY OF COPPER PYRITES. 495 
 
 analysis ; and a quantity of potassium chlorate equal to a 
 small teaspoonful was added to the ore.) Invert a small 
 glass funnel with bent stem in the dish above the pyrites, 
 and pour upon the latter rather more ordinary strong 
 nitric acid than would be sufficient to completely cover 
 the powder. Place the dish upon a water-bath, and, from 
 time to time, throw into it small quantities of potassium 
 chlorate. The doses of the chlorate must be repeated at 
 frequent intervals, until free sulphur can no longer be seen 
 in the dish. If need be, add nitric acid, also, from time to 
 time, to replace that lost by evaporation. 
 
 As a general rule, it is safer and more convenient to 
 heat the mixture on a water-bath than upon sand, though 
 the oxidation of sulphur can be effected more easily and 
 quickly when the mixture of nitric acid and chlorate is 
 heated to actual boiling than at the temperature obtain- 
 able by means of a water-bath. When the last particles 
 of sulphur have been destroyed, remove the inverted funnel 
 from the dish, rinse it with water, and collect the rinsings 
 in a beaker by themselves. Allow the liquid in the evapo- 
 rating-dish to become cold, pour upon it a quantity of 
 ordinary strong hydrochloric acid rather larger than the 
 quantity of nitric acid taken at first, evaporate the mixed 
 solution to dryness, and heat the dry residue to render 
 silica insoluble, in case any silica be present. 
 
 Pour water upon the cold residue, and, without filter- 
 ing the liquor, wash the contents of the dish into the 
 beaker which contains the rinsings of the funnel. Heat 
 the liquid in the beaker nearly to boiling, add to it about 
 25 c.c. of a strong aqueous solution of ferrous sulphate 
 slightly acidulated with sulphuric acid, and keep the mix- 
 ture at a temperature near boiling during four or five 
 minutes, in order to destroy the small quantity of nitric 
 .acid which may have escaped decomposition in spite of 
 the dry evaporation with hydrochloric acid. 
 
 The ferrous salt seldom acts instantaneously, but the 
 reducing action proceeds rapidly and satisfactorily when 
 once begun. If need be, add more of the ferrous solution, 
 little by little, until the entire contents of the beaker 
 
496 THE ASSAY OF COPPER. 
 
 become dark-coloured or black, and no more gas is dis- 
 engaged. 
 
 In order to be sure that all the nitric acid has been 
 reduced, it is as well, after the mixture of liquid and solu- 
 tion of ferrous sulphate has been duly heated, to place 
 a drop of the mixture upon porcelain, and test it with 
 potassium ferrocyanide. In general, however, the colora- 
 tion of the liquid in the beaker, due to the formation of 
 nitrous or hyponitric acid, will be a sufficient indication 
 that the sulphate of iron has done its work. The nitrous 
 fumes quickly disappear from the liquid at a subsequent 
 stage of operations when metallic iron is immersed in the 
 solution. 
 
 When enough of the ferrous sulphate has been added, 
 filter the mixed solution into a wide beaker, precipitate the 
 copper in the metallic state upon a sheet of iron in the 
 usual way, and ignite the copper in a porcelain crucible, 
 in a current of hydrogen, before weighing it. 
 
 By means of the ferrous salt, the last traces of nitric 
 acid may be got rid of far more quickly, conveniently, and 
 certainly than by the old system of evaporating the pyrites 
 solution with several successive portions of hydrochloric 
 acid. By treating the pyrites with potassium chlorate and 
 nitric acid it is easy to oxidise and dissolve every particle 
 of the sulphur in the mineral, so that no portion of the 
 latter can escape decomposition by becoming enveloped 
 in free sulphur. When aqua regia is used, on the other 
 hand, or a mixture of potassium chlorate and hydrochloric 
 acid, a certain proportion of sulphur almost invariably 
 remains undissolved, and might easily enclose portions of 
 the mineral, so as to prevent them from the solvent action 
 of the acids. 
 
 For the Estimation of Copper and Sulphur in Copper 
 Pyrites, E. Fresenius (' Zeitschrift fur Anal, Chemie,' 1877, 
 p. 355) proceeds as follows : 
 
 ' After drying at 100 C., and carefully preparing the 
 sample, he takes for the estimation of the copper 5 
 grammes pyrites, heats with 6 to 7 c.c. hydrochloric acid 
 (sp. gr. 1'17), adding gradually nitric acid of sp. gr. 1-37, till 
 
SULPHUR IX COPPER PYRITES. 497 
 
 no more action ensues, and then digests for some hours 
 at a gentle heat. The contents of the flask are poured 
 into a porcelain capsule, the flask is twice rinsed, out with 
 10 c.c. hydrochloric acid (sp. gr. 1*12) into the capsule, 
 and is then set aside. The contents, of the capsule are 
 then evaporated almost to dry ness on the water-bath, 
 20 c.c. hydrochloric acid of sp. gr. 1/12 are added, heated, 
 mixed with water, and filtered into a boiling-flask holding 
 500 c.c. The solution-flask is also rinsed upon the filter 
 with water. The filtrate is dried, incinerated in a porcelain 
 crucible, and the residue (in part lead sulphate) is treated 
 with 1 c.c. aqua regia, evaporated to dryness, the residue 
 treated with 5 c.c. hydrochloric acid (1*12 sp. gr.) slightly 
 diluted, and the solution, which may contain a little 
 copper, is filtered into the main solution. The solution is 
 then precipitated with sulphuretted hydrogen at 70 C., 
 the precipitate is filtered, washed, dried, mixed with the 
 thoroughly burnt ash of the filter, heated with 5 c.c. 
 nitric acid (of sp. gr. 1-2), filtered, and washed. The filter 
 is incinerated, the small residue is again heated with 2 c.c. 
 of the same nitric acid, diluted, filtered to the main solu- 
 tion, and washed. The solution is mixed with 12 c.c. 
 dilute sulphuric acid (1 part acid to 5 water), evaporated 
 to expel nitric acid, filtered to remove lead sulphate, 
 washing with water acidulated with sulphuric acid. The 
 copper is then precipitated with sulphuretted hydrogen 
 at 70 C., the precipitate washed, dried, ignited along 
 with the ash of the filter in a current of hydrogen, and 
 weighed as copper sulphide. 
 
 6 For the estimation of sulphur the author fuses 
 ^ grm. of the sample with 10 parts of a mixture of 
 2 parts sodium carbonate and 1 part potassium nitrate, 
 and estimates the sulphuric acid formed. For burnt ores 
 he uses 4 parts sodium carbonate to 1 part of potassium 
 nitrate.' 
 
 Professor Chapman, of Toronto, gives the following 
 directions for the detection of minute traces of copper in 
 iron pyrites and other bodies : 
 
 Although an exceedingly small percentage of copper 
 
 K K 
 
498 THE ASSAY OF COPPER, 
 
 may be detected in blowpipe experiments by the reducing 
 process, as well as by the azure-blue coloration of the 
 flame when the test matter is moistened with hydrochloric 
 acid, these methods fail in certain extreme cases to give 
 satisfactory results. It often happens that veins of iron 
 pyrites lead at greater depths to copper pyrites. In this case, 
 according to the experience of the writer, the iron pyrites 
 wi]l almost invariably hold minute traces of copper. Hence 
 the desirability, on exploring expeditions more especially, 
 of some ready test by which, without the necessity of 
 employing acids or other bulky and difficultly portable 
 reagents, these traces of copper may be detected.* The 
 following simple method will be found to answer the pur- 
 pose : The test substance, in powder, must first be roasted 
 on charcoal, or, better, on a fragment of porcelain,f in 
 order to drive off the sulphur. A small portion of the 
 roasted ore is then to be fused on platinum wire with 
 phosphor-salt ; and some potassium bisulphate is to be 
 added to the glass (without this being removed from the 
 wire) in two or three successive portions, or until the glass 
 becomes more or less saturated. This effected, the bead 
 is to be shaken off the platinum loop into a small capsule, 
 and treated with boiling water, by which either the whole 
 or the greater part will be dissolved ; and the solution is 
 finally to be tested with a small fragment of potassium 
 ferrocyanide. If copper be present in more than traces, 
 this reagent, it is well known, will produce a deep red 
 precipitate. If the copper be present in smaller quantity 
 that is, in exceedingly minute traces the precipitate 
 
 * In blowpipe practice as far, at least, as this is possible the operator 
 should make it an essential aim to render himself independent of the use of 
 mineral acids and other liquids and inconvenient reagents of a similar 
 character. If these reagents cannot be dispensed with altogether, their use, 
 by improved processes, may be greatly limited. 
 
 t In the roasting of metallic sulphides, &c., the writer has employed, for 
 some years, small fragments of Berlin or Meissen porcelain, such as result 
 from the breakage of crucibles and other vessels of that material. The test 
 substance is crushed to powder, moistened slightly, and spread over the sur- 
 face of the porcelain ; and when the operation is finished, the powder is easily 
 scraped off by the point of a knife-blade or small steel spatula. In roasting 
 operations, rarely more than a dull red heat is required ; but these porcelain 
 fragments may be rendered white-hot, if such be necessary, without risk of 
 fracture. 
 
ESTIMATION OF ARSENIC IN COPPER. 499 
 
 will be brown or brownish- black ; and if copper be en- 
 tirely absent, the precipitate will be blue or green assum- 
 ing, of course, that iron pyrites or some other ferruginous 
 substance is operated upon. In this experiment the pre- 
 liminary fusion with phosphor-salt greatly facilitates the 
 after solution of the substance in potassium bisulphate. 
 In some instances, indeed, no solution takes place if this 
 preliminary treatment with phosphor-salt be omitted. 
 
 Estimation of Arsenic in Copper. 
 
 The estimation of the small quantity of arsenic always 
 present in commercial copper, or the separation of a very 
 small quantity of arsenic from a large amount of copper, 
 is a matter of considerable difficulty, as the ordinary 
 methods of separation fail to give accurate results. 
 
 Having had a very extensive experience in the analysis 
 of copper, and knowing the extraordinary discrepancies 
 which occur in analyses of the same sample of copper by 
 different chemists, which can only arise from the use of 
 imperfect methods, Mr. A. Humboldt Sexton proposes 
 method which he has found to give quite accurate 
 results. 
 
 The copper is dissolved in nitric acid, a small quantity 
 of solution of ferric nitrate added, the solution nearly 
 neutralised with sodium hydrate (not ammonia), and excess 
 of sodium acetate added. The solution is then heated to 
 boiling, and filtered as rapidly as possible : the precipitate 
 after being well washed is dissolved in hydrochloric acid, 
 the solution made alkaline with ammonia and saturated 
 with sulphuretted hydrogen, and filtered from the precipi- 
 tated iron sulphide. The filtrate is acidified with hydro- 
 chloric acid and allowed to stand in a warm place for 
 some time. The arsenic and antimony sulphides are 
 filtered off, dried at 100 C., the precipitates removed 
 completely from the paper into a small beaker, treated 
 with red fuming nitric acid, a few drops of hydrochloric 
 acid being added as soon as the action has ceased. It is 
 then diluted, filtered, the arsenic precipitated as ammonia- 
 
500 THE ASSAY OF COPPER. 
 
 magnesium arseniate, and weighed as usual. If the preci- 
 pitated sulphides cannot be perfectly removed from the 
 filter-paper, the paper must be treated with nitro-hydro- 
 chloric acid, filtered, and the filtrate added to the nitric 
 acid solution. 
 
 This method is very accurate, and each stage has been 
 carefully experimented upon. It requires, however, some 
 special precautions. 
 
 When the sodium acetate is added, the colour of the 
 solution should change from pale blue to dark green ; this 
 shows that the solution has been sufficiently neutralised. 
 The beaker must be removed from the heat immediately 
 the solution begins to boil ; if the solution be left boiling 
 (sometimes when it is not), a greenish white precipitate 
 of basic copper acetate- falls. This can generally be 
 removed by the addition of a few drops of hydrochloric 
 acid, but in cases where it has separated on the surface of 
 the beaker, or where it will not readily dissolve, it is best 
 to throw out the solution and commence again. 
 
 This is very troublesome to those using this method for 
 the first time, but after a little experience has been gained 
 it very rarely happens. 
 
 The precipitate should have the dark red colour of 
 ferric acetate ; if it is paler it is due either to there not 
 being sufficient iron, or to the co-precipitation of some 
 basic copper acetate. The filtrate should be blue or 
 pale green : sometimes it is dark green and turbid, from 
 the presence of iron acetate carried through the filter ; in 
 that case the first portions must be passed through the 
 filter again. 
 
 The precipitate must be washed till it is free from 
 copper, and when it is dissolved in hydrochloric acid the 
 solution must have the yellow colour of ferric chloride. 
 If it is at all green, the solution must be neutralised, a 
 little more sodium acetate added, and the iron and arsenic 
 re-precipitated. 
 
 Mr. Sexton has made a larger number of experiments 
 in order to ascertain the amount of iron necessary for 
 complete precipitation. With equal quantities of iron 
 
ESTIMATION OF ARSENIC IN COPPER. 501 
 
 and arsenic, a small quantity of arsenic remained in solu- 
 tion, and the iron-arsenic precipitate was of a pale colour. 
 With 1-5 parts of iron to 1 of arsenic the precipitation 
 was complete. In order to make sure, it is well to add 
 about twice as much iron as it is expected there is arsenic 
 present. Then, even if a little iron remains unprecipitated, 
 all the arsenic will be thrown down. 
 
 Since copper sulphide retains so much arsenic, it might 
 be expected that iron sulphide would act in a similar 
 manner, but it does not ; if there be no copper present 
 the precipitate is quite free from arsenic, but if copper is 
 present a considerable quantity of arsenic may be retained. 
 Hence the importance of thoroughly washing the acetate 
 precipitate, and re -precipitating it if necessary. 
 
502 
 
 CHAPTER XI. 
 
 THE ASSAY OF LEAD. 
 
 ALL minerals and substances containing lead may, for the 
 purposes of assay by the dry way, be divided into four 
 classes : 
 
 Class I. comprises sulphides, antimonial or otherwise 
 (galena, &c.) 
 
 Class II. includes all plumbiferous substances contain- 
 ing neither sulphur nor arsenic, or mere traces only of 
 these elements (litharge, minium, lead carbonate, native 
 and artificial, lead fume, cupel bottoms, furnace hearths, 
 lead slag, &c.) 
 
 Class III. comprises all substances into whose com- 
 position either sulphuric, arsenic, chromic, or phosphoric 
 acid, or a mixture of either, enters (pyromorphite, wolfram- 
 ite, &c.) 
 
 Class IV. Alloys of lead. 
 
 CLASS I. 
 
 Before describing the different modes of assaying sub- 
 stances of this class, it will be as well to pass in review the 
 action of various reagents on sulphides of lead, in order 
 that the rationale of the assay of those ores may be better 
 appreciated. 
 
 Action of Oxygen. If galena be roasted at a very 
 gentle temperature, care being taken to avoid fusion, it- 
 will be converted into a mixture of lead oxide and lead 
 sulphate, with evolution of sulphurous acid, thus: 
 
 2(PbS) + 70 =PbO + PbO,S0 3 + S0 2 . 
 Action of Metallic Iron. This metal completely and 
 
ALKALIES AND ALKALINE CARBONATES. 503 
 
 readily decomposes lead sulphide, giving metallic lead in 
 a pure state, thus : 
 
 On the one side we have lead sulphide and metallic iron, 
 on the other metallic lead and iron sulphide. 
 
 The Alkalies and Alkaline Carbonates decompose lead 
 sulphide, but only partially ; pure lead is separated, and 
 at the same time a very fusible grey slag is formed, which 
 contains an alkaline sulphate and a compound of lead sul- 
 phide and an alkaline sulphide. A certain proportion of 
 the alkali is reduced by the sulphur, which is converted 
 into sulphuric acid, so that no lead oxide is produced. 
 This reaction may be thus expressed : 
 
 7(PbS) + 4(K 2 0)=4Pb + K 2 0,S0 3 + 3(PbS,K 2 S). 
 
 Potassium Nitrate completely decomposes lead sul- 
 phide, with the reduction of metallic lead and formation 
 of potassium sulphate and sulphurous acid, thus : 
 
 If the nitre be in excess, the lead will be oxidised in pro- 
 portion to the excess present ; and if there be a sufficiency 
 added, no metallic lead at all will be produced. 
 
 Argol. The presence of carbonaceous matter much 
 favours the decomposition of galena, by determining the 
 reduction of a larger quantity of potassium, and thereby 
 the formation of a larger quantity of alkaline sulphide. 
 With 4 parts of argol to 1 part of sulphide, 80 parts of 
 lead are reduced. If the reaction were complete, the 
 decomposition would be as follows : 
 
 For the reactions of lead oxide (litharge) and lead sul- 
 phate on sulphide of lead, see pages 187 and 188. 
 
 From the reactions above given, it will be seen that 
 there are many substances capable of completely reducing 
 the lead from its sulphide, and yet few can be used safely 
 
504 THE ASSAY OF LEAD. 
 
 with any advantage, as so to use them would imply a 
 knowledge of how much sulphur and lead were in the ore 
 to be assayed, in order to tell the precise quantity of either 
 of the reagents required ; for it is evident that if either 
 more or less of some were added, a faulty result would be 
 the consequence : so that some systematic mode of assay, 
 which may be suitable for all classes of galena, whether 
 mixed with other sulphides or with gangue, must be 
 contrived. To facilitate this we now proceed to give an 
 outline of the methods generally adopted in the assay of 
 lead ores by various processes. 
 
 1. FUSION WITH POTASSIUM CARBONATE. 
 
 This plan is used at the Oberhartz, and described by 
 Kerl as follows : 
 
 One centner of the very finely pulverised assay sub- 
 stance is weighed out, mixed with three to four times its 
 weight of pure, dry, and finely pulverised potassium car- 
 bonate, and covered over, in a small clay crucible (fig. 61), 
 with a layer of decrepitated sodium chloride about one- 
 fourth of an inch thick. The assays thus prepared are 
 placed in the thoroughly heated muffle of a large assay 
 furnace (figs. 23, 24) having a strong draught. They re- 
 main in the highest temperature of the furnace, with the 
 mouth of the muffle closed with glowing coals, till they 
 have come into perfect fusion (about twenty to thirty 
 minutes). The draught opening is then closed, and at the 
 same time the muffle opened, until the temperature has 
 fallen so far that the crucibles appear brownish red, and 
 the vapours above them have greatly diminished, or have 
 disappeared. At this heat the crucibles whose contents 
 must, however, always remain in perfect fusion are main- 
 tained, according to the fusibility and composition of the 
 assay sample, and the draught of the furnace for a longer 
 or snorter time (ten to twenty-five, generally ten minutes). 
 This period, during which the heat is allowed to remain 
 low, is called the cooling of the assay. 
 
 The furnace is now again brought back to its first 
 
FUSIOX WITH POTASSIUM CAEBONATE. 505 
 
 temperature by completely opening the draught and 
 closing the muffle. Ten to fifteen minutes of this last 
 heating are in most cases sufficient. Only poor ores, &c., 
 which contain also a pretty large quantity of arsenic, or 
 of iron, zinc, and copper sulphides, are allowed to continue 
 hot five to ten minutes longer. 
 
 If many assays are to be made, it will be found advan- 
 tageous to mix those which contain larger quantities of 
 foreign sulphides, or, by reason of their earthy contents, 
 are difficultly fusible, with more or less borax ; or, instead 
 of this, to place them in the back and hotter part of the 
 muffle, while those that are very rich in lead and easily 
 fusible are placed in front, since the latter will be hot 
 enough here, and more easily reached by the air than 
 those deeper in the muffle. 
 
 The crucibles, when cold, are broken, the lead buttons 
 obtained are freed from all adhering slag or substance of 
 the crucible, and if the assay were otherwise successful 
 their weight found. The assays should not be too rapidly 
 cooled, because the slag is thus easily cracked, and the 
 still half-fluid button lying below is apt to be broken into 
 several pieces. 
 
 In a successful assay, the lead melted together to a 
 button deports itself under the hammer and knife like pure 
 lead, and possesses also its colour. If the slag shows, upon 
 its surface of separation from the metallic button, lead- 
 grey spots with metallic lustre, it will generally also be 
 found that a thin layer of not completely decomposed 
 glistening lead sulphide or subsulphide has at the same 
 time deposited itself upon the button. This layer, if the 
 above appearance presents itself in a high degree, can be 
 rubbed off or removed in fine scales. The lead button 
 itself then shows upon its surface a high metallic lustre, 
 which does not have the colour of pure lead, but a darker 
 and blackish hue. Assays of this kind are to be rejected ; 
 they have not been allowed to remain cool long enough, 
 or they have in the process become too cold: they give 
 the amount of lead too low, and often very considerably 
 so. In assays which have stood too long in the furnace 
 
506 THE ASSAY OF LEAD. 
 
 in the last fusing heat, a very bright button of lead is also 
 found ; but here the layer of undecomposed lead sulphide 
 is wanting, as also the glistening spots on the surface of 
 the slag surrounding the button. If the influence of the 
 heat and air continues too long, then, besides a loss through 
 volatilisation of the lead, a slagging of the lead oxide may 
 take place. A button that is brittle, laminated, and bril- 
 liantly white in the fracture, indicates an insufficiency of 
 flux, or the presence of antimony and arsenic. In successful 
 assays the lead button generally has a bluish appearance, 
 which, although not dull, is at the same time not strongly 
 brilliant. The slag must be completely homogeneous, and 
 must have settled down uniformly towards the bottom of 
 the crucible, so that it does not stick in a thick layer to 
 the upper part of the sides of the crucible. It shows by 
 this that it has been in proper fusion. It must have 
 covered over the button in a thick layer (about one-fourth 
 of an inch thick). The sodium chloride covering, or a 
 more or less colourless slag that is formed containing 
 sodium chloride and potassium carbonate, overlies in a 
 still thicker layer the true dark-coloured slag containing 
 the foreign metallic oxides. A porous slag containing 
 metallic globules indicates a too small quantity of flux or 
 too low a temperature ; a brilliant vitreous slag, too high 
 a temperature and a slagging of lead. An assay and its 
 duplicate must, moreover, give equal results. 
 
 Lead matte and lead fume are smelted, with the addi- 
 tion of borax and coal-dust, with potassium carbonate, and 
 with the first the heat is allowed to last somewhat longer 
 (perhaps to three-quarters of an hour) than with ores. 
 The potassium carbonate assay gives for lead matte, with its 
 not inconsiderable lead contents (30 per cent, and over), 
 pretty satisfactory results. 
 
 The theory of this lead assay appears from the fol- 
 lowing. 
 
 If perfectly pure galena is intimately mixed with three 
 or four times its weight of good dry potassium carbonate, 
 placed in a clay retort, and this so arranged in the muffle 
 of the assay furnace that its neck projects from the mouth 
 
FUSION WITH POTASSIUM CARBONATE. 507 
 
 of the muffle, while in the opening of the neck a glass tube 
 is closely fitted, which goes into a receiver, from which it 
 is further prolonged in a second tube, it will be observed 
 that at first only a little water collects in the receiver, pro- 
 ceeding from the small quantity of moisture always present 
 in the potassium carbonate. Later, with an incipient red 
 heat in the retort, a gas is disengaged, which upon closer 
 investigation proves to be pure carbonic acid gas, i.e. free 
 from sulphurous acid. The disengagement of gas becomes 
 more active with a stronger red heat, without yielding 
 gases of different composition, but ceases again after a 
 while. In order to obtain assurance of a complete decom- 
 position, the retort may be kept for an hour at a very 
 strong red heat. After the cooling and breaking of the 
 retort, some pure lead oxide and carbonate is found 
 deposited in the neck of it, then a pure lead button upon 
 the bottom, and over this a brown slag, free from little 
 globules of lead. It consists in by far the greatest part 
 of potassium sulphide and still undecomposed potassium 
 carbonate, but also in small part of potassium silicate de- 
 rived from the silica of the retort. If this slag is treated 
 with water till nothing further will dissolve, the substances 
 named can be easily shown to exist in the solution. The 
 solution is colourless, and when supersaturated with acids 
 disengages sulphuretted hydrogen, but throws down no 
 sulphur. In the treatment of the slag with water, lead 
 sulphide remains behind in black flocks, even the superfi- 
 cial character of which shows that it is not undecomposed 
 galena, but lead sulphide separated from a chemical com- 
 bination. 
 
 If the brown slag from the retort is placed in a small 
 uncovered crucible and brought back into the hot muffle 
 of the assay furnace and melted, then after some time, 
 whether the slag was covered with sodium chloride or not, 
 a button of lead again separates at the bottom of the cru- 
 cible, and the brown slag now shows itself decolourised. 
 If the crucible is removed from the furnace too soon, 
 only the upper layer of slag is decolourised, and that lying 
 below is still completely unchanged. The decolourised slag 
 
508 THE ASSAY OF LEAD. 
 
 consists of potassium carbonate and sulphate, and no 
 longer contains any trace of potassium sulphide. 
 
 In the above-described lead assay, the process in the 
 strong preliminary heat proceeds as in the retort, i.e. the 
 potash of the potassium carbonate is reduced to potassium, 
 while it yields its oxygen to the sulphur of the galena and 
 with it forms sulphuric acid ; the liberated potassium takes 
 up sulphur from another portion of galena, forming potas- 
 sium sulphide. The galena would now in this double way 
 soon lose all its sulphur, if a combination a sulphur salt 
 of potassium sulphide with lead sulphide did not form, 
 which resists all further action of the potassium carbonate 
 [4 (K 2 0,C0 2 ) + 7PbS = 4Pb + 3 (K 2 S,PbS) +K 2 0,S0 3 + 4COJ. 
 The carbonic acid of the thus decomposed potassium car- 
 bonate escapes together with that set free by the sulphuric 
 acid formed, and causes a puffing up of the mass, by 
 which globules of lead already separated are raised up 
 with it, and may perhaps remain with some of the slag 
 sticking to the upper crucible walls. They would here 
 oxidise and produce yellow spots. The covering of sodium 
 chloride is designed to guard against loss of lead in this 
 and similar ways. It serves in a certain manner to rinse 
 down the sides of the crucible. 
 
 The atmospheric oxygen, in the open crucible, is not 
 entirely excluded by the covering of sodium chloride. In 
 the cooling of the assay, it oxidises the sulphur salt con- 
 tained in the upper part of the slag, forming potassium 
 sulphate and a portion of lead sulphate. The latter, 
 during the last high heat, decomposes the lead sulphide 
 still remaining in the slag, in such a way as to produce 
 metallic lead (PbS + PbO,S0 3 = 2Pb + 2S0 2 ). The reduced 
 particles of lead separate well from the slag thus rendered 
 thinly fluid. Mattes must be allowed to cool longer than 
 ores. 
 
 The potassium carbonate assay presupposes in general 
 great practice and close attention on the part of the 
 assayer ; and moreover, if one wishes to find the correct 
 value at once, without fruitless preliminary examinations, 
 and without the necessity of repeating the assay, a general 
 
FUSION WITH POTASSIUM CARBONATE. 509 
 
 knowledge of the constituents of the assay sample, so far, 
 for example, as this can be obtained by the aid of minera- 
 logy, is necessary. The assay after this method, which 
 requires but little preparation, can only be conducted in 
 the muffle furnace, but then in pretty large number (as 
 many as fifty at once). For its success it is indispensably 
 necessary that the cooling of the assay be allowed and 
 stopped again at the right time and in the proper degree. 
 If it is allowed to cool too long, too much sulphate of lead 
 is formed in proportion to the lead sulphide still present 
 in the slag, and in the last heating up, by the action 
 of the two upon each other, easily scorifiable oxide of 
 lead is produced (PbS + 3PbO,S0 3 =4PbO + 4S0 2 ). If 
 the cooling is too soon interrupted, only a small part of 
 the lead sulphide in the sulphur salt is oxidised, and, 
 by the action of the oxidised portion upon the lead 
 sulphide, lead sub-sulphide is produced, which either 
 remains in the slag or settles upon the lead button 
 (2PbS + PbO,S0 3 = Pb 2 S + 2Pb + 2S0 2 ). Experience gives 
 the only means at hand to guide us here, but leaves 
 us easily in the lurch, so that the result of the assay 
 becomes more doubtful than in some of the methods- 
 hereafter described. 
 
 With substances containing antimony this assay deserves 
 the preference over the others, since most of the antimony 
 remains in the slag in the state of sulphide and oxide. An 
 addition of saltpetre works advantageously. Arsenic and 
 arsenic sulphide mostly go off in fumes during the smelt- 
 ing, but nevertheless always cause the formation of a 
 brittle metallic button. Copper sulphide remains in great 
 part in the slag, but a part of the copper is desulphurised 
 and goes into the lead. If the quantity of copper present 
 is very considerable, the button of metal may be considered 
 as black copper, and refined, and the loss thereby occur- 
 ring reckoned as lead. 
 
 Iron protosulphide, which occurs, for example, in lead 
 matte, is decomposed by potassium carbonate, forming 
 metallic iron, which desulphurises the galena. Iron pyrites , 
 on the other hand, occasions the forming of a large quan- 
 
510 THE ASSAY OF LEAD. 
 
 tity of potassium sulphide and, in consequence of this, of 
 a sulphur salt. 
 
 It follows, therefore, from the above, that ores which 
 contain much foreign sulphides are not suited to this 
 method of assaying, since they cause the production of a 
 large amount of potassium sulphide, which always retains 
 lead sulphide. By an addition of saltpetre to the potas- 
 sium carbonate these sulphides may, indeed, be partially 
 decomposed : only an oxidation of the lead is apt to be 
 produced, as well as a mechanical loss by the violent 
 action of the saltpetre. 
 
 From pure galena, by the potassium carbonate assay, 
 80 per cent, of lead at most can be obtained. Calcined 
 sodium carbonate is inferior to potassium carbonate as a 
 desulphurising agent, and always yields a few per cent, 
 less lead than the latter. According to .Phillips, 75 to 77 
 per cent, of lead is obtained from galena with sodium car- 
 bonate. With potassium cyanide, under certain circum- 
 stances, the same result can be obtained as with potassium 
 carbonate, and it does not require so high nor so long- 
 continued a temperature ; still it offers no real advantage 
 over potassium carbonate. An addition of 30 to 35 per 
 cent, of saltpetre to an assay, with which ten parts of 
 sodium carbonate are used, promotes, indeed, the desul- 
 phurising of the lead, but also increases its loss. 
 
 At the Oberhartz smelting-house the lead button is 
 weighed out to pounds, and a difference of five pounds is 
 allowed between different assayers. It is also a custom, 
 though not a correct one, to allow as many pounds differ- 
 ence as there are tens of pounds in the weight of the lead 
 button obtained. Thus, with a lead contents of thirty 
 and seventy pounds, the difference in the separate assays 
 might amount to three and seven pounds respectively. 
 
 2. FUSION WITH BLACK FLUX. 
 
 A modification of the preceding method of assaying, 
 which is sometimes employed, consists in using, instead of 
 the potassium carbonate, an equal quantity of black flux, 
 
FUSION WITH METALLIC IRON. 511 
 
 or indeed of argol, or in mixing a few per cent, of pow- 
 dered charcoal or flour with the potassium carbonate, or 
 in replacing it in part by argol. Too great an addition of 
 carbon diminishes the fusibility of the mass, and hinders 
 the flowing together of the separated particles of lead. By 
 using argol the operation lasts longer, because the mass 
 remains pasty until most of the tartaric acid has been 
 decomposed ; but a greater product of lead is obtained. 
 The chemical reaction during the operation is thereby 
 modified so that the carbon of the black flux exerts an 
 influence upon the potash, and partially reduces it to 
 potassium ; the potassium, thus set free, works now, as 
 before, upon the galena. The latter is thus, without the 
 influence of the air, more completely decomposed than by 
 potassium carbonate alone, and the smelting is, therefore, 
 conducted in covered crucibles (fig. 61) in the wind fur- 
 nace. But since there is also potassium sulphide formed, 
 and this dissolves lead sulphide, it is more advisable, for 
 the completest possible separation of the lead, to perform 
 the smelting in open crucibles in the mulfle, in order to 
 allow the atmospheric oxygen to work at the same time 
 on the assay. The product of lead from pure galena does 
 not generally exceed 76 to 79 per cent. 
 
 At the Victor-Frederick smelting works in the Hartz, 
 one centner (= one hundred and fourteen assay pounds) 
 of galena is mixed with three or four times as much black 
 flux, and with pyritic ores ten pounds of borax-glass are 
 added. The mixture is covered with sodium chloride, 
 heated for about twenty-five minutes in the muffle furnace 
 with a charcoal fire, and then, after the mouth of the 
 muffle has been opened for about five minutes, taken out 
 of the furnace. 
 
 3. FUSION WITH METALLIC IRON. 
 
 Schlutter and many of the older assayers were aware 
 that iron would desulphurise galena, and ever after advised 
 the addition of a certain quantity of that metal to the 
 different fluxes which they used in lead assays ; but it was 
 
512 THE ASSAY OF LEAD. 
 
 at the practical School of Mines, at Montiers, that iron was 
 first employed alone. 
 
 The process is extremely convenient and easy of exe- 
 cution ; it always succeeds, and requires no troublesome 
 attention. The fusion takes place quietly, without frothing 
 or bubbling ; and as the whole substance employed requires 
 but little space, very small pots may be employed, or a 
 very large quantity assayed. But this process can only be 
 employed for pure galenas, or those which contain at most 
 a few per cent, of gangue. 
 
 When galena is heated with iron, the metal is trans- 
 formed into protosulphide, whence it follows, that to de- 
 sulphurise galena 22 '6 per cent, is required ; but expe- 
 rience has shown that it is better to employ a little more, 
 and 30 per cent, can be used without inconvenience. The 
 iron employed ought to be in the state of filings, or wire 
 cut very small. The mixture is placed in a crucible, 
 which is three-fourths filled ; the whole is covered with a 
 layer of salt, sodium carbonate, or black flux, and exposed 
 to a full red heat. After the flux is perfectly fused, the 
 pot may be cooled and broken, and a button is obtained, 
 which at first sight has a homogeneous aspect, but on 
 being struck with the hammer separates into two distinct 
 parts. The lower part is ductile lead : the upper, a very 
 brittle matte, of a deep bronze colour, and slightly mag- 
 netic. Pure galena yields, by this process, 72 to 79 per 
 cent, of lead, so that there is a considerable loss, which 
 loss is entirely due to volatilisation. Berthier says that 
 it does not appear possible to avoid this loss, which 
 amounts from 6 to 13 per cent., giving as a reason that it 
 is probable galena begins to sublime before it arrives at 
 the proper heat for decomposition. 
 
 Antimonial galenas, or galenas mixed with iron pyrites, 
 may be assayed in the same manner ; but then a sufficiency 
 of iron must be added to reduce the antimony to the 
 metallic state, as well as to reduce the iron pyrites to the 
 minimum of sulphurisation. If the galena be mixed with 
 blende, the greater portion remains in the slag, because it 
 is only decomposed by iron at a very high temperature. 
 
FUSION WITH METALLIC IRON. 513 
 
 Blende being infusible by itself, much diminishes the 
 fusibility of the mattes produced ; and if it exists in very 
 large quantity, it can even hinder their complete fusion ; in 
 which case some iron protosulphide and metallic iron must 
 be added to the assay, to make the slag more fusible. 
 
 All minerals are at a minimum of sulphurisation when 
 existing in mattes from metallurgical works ; therefore 
 much less iron may be used in their assay than if they 
 were pure ores. In very rich lead mattes, in which the 
 lead exists as subsulphide, from 10 to 12 per cent, is 
 sufficient. A small excess of iron may be employed with- 
 out inconvenience ; but if a larger proportion be added 
 than is necessary to execute the desulphurisation, the matte 
 contains some iron in the metallic state, and loses its 
 liquidity, and in consequence retains some globules of 
 lead. 
 
 The usual mode of assaying lead ores (galena) in the 
 lead mills is by a modification of this process : in lieu of 
 placing the ore in an earthen crucible, and adding nails or 
 filings, a given weight of the ore is projected into a red- 
 hot wrought- iron crucible, which is kept in the fire for 
 about a quarter of an hour, or until all the galena seems 
 decomposed. The lead thus reduced is poured into a 
 mould ; and if the scoriaceous matter be not well fused, 
 the iron crucible is returned to the fire and heated still 
 more strongly, and any lead that may be separated is 
 poured into the mould and weighed with the rest. This 
 is a very rude and imperfect process, and gives only 
 tolerable results with pure galenas, but is quite unsatisfac- 
 tory with those containing much earthy matter, as not 
 above half the lead is obtained, owing to volatilisation and 
 exposure to the air, and the loss of globules in the slag. 
 This process succeeds much better when a flux is added ; 
 this may be argol, or sodium carbonate, or a mixture of 
 both (see next process). 
 
 L L 
 
514 THE ASSAY OF LEAD. 
 
 4. FUSION WITH SODIUM CARBONATE OR BLACK FLUX, AND 
 METALLIC IRON. 
 
 When galena is heated with an alkaline flux, out 
 of contact of air, the slag contains a double sulphide of 
 lead and the alkaline metal employed : if iron be thrown 
 into this fused mixture metallic lead separates, and the 
 iron combines with the sulphur formerly combined with the 
 lead, and the slag will contain a double alkaline sulphide, 
 containing iron sulphide instead of lead sulphide, thus : 
 
 PbS + K 2 S + Fe - Pb 2 + FeSK 2 S. 
 
 Any earthy substances the ore may contain will be 
 dissolved by the alkaline flux, without very much impair- 
 ing its fluidity. All these facts being considered, it may 
 be readily seen that the assay of all earthy bodies contain- 
 ing lead sulphide may be made in this manner, with as 
 much accuracy as this method of assay can be capable of. 
 Either sodium carbonate or black flux may be employed 
 as the alkaline reagent, and more of either of those 
 substances must be employed, in proportion to the in- 
 creased quantity of earthy matters the ore contains. Two 
 parts are nearly always more than sufficient for poor 
 ores, and are best for all cases, because an excess of 
 flux does not diminish the yield of lead ; nevertheless it 
 is sometimes convenient to employ, for the latter class, 
 but half a part. As to the iron, it is employed only to 
 separate that part of the lead which has been dissolved in 
 the state of sulphide by the alkali, but not decomposed ; 
 so that much less may be employed than is necessary for 
 the decomposition of the w r hole amount. 
 
 Experiment has shown that the maximum amount of 
 lead from pure galena may be obtained by the use of the 
 following mixtures : 
 
 2 parts of black flux, or sodium carbonate, and 10 to 
 12 of iron. 
 
 1 part of black flux, or sodium carbonate, and 20 of 
 iron 
 
ROASTING AND REDUCING ASSAY. 515 
 
 \ a part of black flux, or sodium carbonate, and from 
 25 to 30 of iron. 
 
 When black flux is employed, and the iron is in the 
 state of filings, it would be inconvenient to employ too 
 much of the latter, especially if the assay were heated 
 very strongly, because the button of lead might be con- 
 taminated with iron ; but when sodium carbonate is used 
 with small iron nails instead of filings, the excess of iron 
 is not inconvenient, but rather useful, because the desul- 
 phurisation is certain to be complete.* 
 
 The following changes take place in both cases. That 
 portion of iron filings mixed with the sodium carbonate 
 which has not been sulphurised is brought to the state of 
 oxide by the carbonic acid of the alkaline carbonate, and 
 remains combined or neglected in the slag ; so that the 
 proportion of iron is never too great, and never becomes 
 mixed with the lead. When black flux is employed, the 
 same oxidation does not take place, on account of the 
 presence of carbonaceous matter, so that the portion of 
 filings not combined with sulphur, and which is merely 
 held in suspension in the flux, passes through it with the 
 globules of lead to the bottom of the crucible ; but if, 
 instead of filings, small nails are employed, they only suffer 
 corrosion at their surface, without change of form or soften- 
 ing, and after the assay are found fixed in the surface of 
 the button of lead, so that they can be detached very 
 readily, and, according to Berthier, without loss of lead. 
 This, however, we have found no easy task, and have 
 always sustained a notable loss. 
 
 5. ROASTING AND REDUCING ASSAY. 
 
 This mode is preferable for ores and substances which 
 contain a considerable quantity of foreign sulphides, or 
 arsenides and antimonides, and a greater or less amount of 
 
 * The French assayers use a piece of plate iron in the shape of a horse- 
 shoe, which is moved about in the melted mass until no more globules of lead 
 attach themselves to it. 
 
 In Germany a mass of iron wire is used. What iron is not consumed by 
 the assay is found still hanging together in a single mass. 
 
 L L 2 
 
516 THE ASSAY OF LEAD. 
 
 earthy matter. It is used in many parts of Germany (at 
 the Unterhartz), and is described by Kerl thus : 
 
 Two assay centners of ore, matte, &c., are heated at first 
 at a low red heat in the muffle, on a roasting dish that has 
 been previously rubbed with chalk. After ten or fifteen 
 minutes they are taken out of the furnace, then again 
 roasted at a moderate temperature for ten or fifteen minutes 
 with frequent turning of the dish. The assay is then once 
 more taken from the furnace, allowed to cool, rubbed up 
 in an agate mortar, and again roasted for half an hour, 
 whereupon it is taken out of the furnace ; tallow is added 
 while it yet glows, and it is again brought to a strong red 
 heat. The rubbing up and calcining with tallow are 
 repeated several times more, and when afterwards the 
 assays have been exposed for two hours continuously to a 
 strong red heat, with the mouth of the muffle almost en- 
 tirely closed, if no more sulphurous acid vapours escape, 
 the roasting is considered as finished. This lasts from six 
 to twelve hours. The roasted sample is then portioned 
 out with the balance, each portion mixed with three or 
 four parts of black flux and an equal quantity of borax 
 and glass, placed in a small crucible covered with sodium 
 chloride, furnished with a little piece of coal as a cover, 
 and smelted in the wind furnace for about a quarter of an 
 hour after the fire is well ignited. Assays that have 
 worked well give nearly equal malleable buttons that do 
 not contain matte, and a black uniformly fused slag. 
 
 The purpose of the roasting is to convert the metallic 
 sulphides, arsenides, and antirnonides into oxides. But 
 since, in the process, sulphates, antimoniates, and arseniates 
 are produced, we seek to destroy these by repeated cal- 
 cining with tallow (see above), instead of an intermixture 
 of coal-dust or flour. By melting the roasted assay with 
 its charge at not too high a temperature, the lead oxide is 
 reduced, and the foreign oxides and earths contained in 
 the sample are, by the aid of the potash in the black flux, 
 as well as of the borax and glass, slagged off. If sulphates 
 or sulphides have remained behind in the roasted ore, 
 they will in the smelting be partially desulphurised by the 
 
ASSAY WITH BLACK FLUX. 517 
 
 action of the oxides, especially of the iron oxide. An 
 addition of metallic iron would in this respect be advan- 
 tageous. 
 
 The roasting is a lengthy process, and one which causes 
 a not unimportant loss of lead. If it is not done tho- 
 roughly, then in the reduction smelting, sulphur salts 
 are formed, which always retain lead, as also a plumbi- 
 ferous matte which surrounds the lead button. By the 
 use of too high a temperature in the smelting a great part 
 of the foreign oxides is reduced, and the lead becomes 
 contaminated. The reduction, however, cannot be entirely 
 avoided, even with a rightly conducted temperature. 
 
 Galena melts less easily than metallic lead if the air 
 is excluded ; but is much more volatile than the latter, 
 and is decomposed by fusion into a higher sulphide 
 which is volatile, and a lower one (Pb 2 S) which remains 
 as a residue. Galena by roasting gives a mixture of lead 
 oxide and sulphate, from which last the sulphuric acid 
 cannot be separated, even at a fusing temperature. Lead 
 sulphate becomes soft by heat, fuses at a bright white 
 heat, and is converted by carbon, with a considerable loss 
 of lead through volatilisation, into lead oxide, metallic lead, 
 or lead subsulphide, acording to the quantity of carbon 
 used and the temperature employed. With lead oxide 
 the sulphate easily fuses together. 
 
 ADDITIONAL REMARKS ON THE LEAD ASSAY. 
 
 Comparison of the Different Methods for the Docimastic Esti- 
 mation of Lead in their Application to Various Products. 
 
 Markus has made the following comparative experi- 
 ments with the methods of assaying lead ores most in use 
 at the Austrian smelting works at Joachimsthal. 
 
 a. Assay with Slack Flux and Iron. One assay centner 
 (5*7 grammes) of the finely rubbed, sifted, and dried assay 
 substance was mixed with two assay centners of black flux, 
 made of sixteen saltpetre and forty argol, and sixty 
 of borax-glass in a mixing capsule, and put into a clay 
 crucible, on the bottom of which a piece of thick iron wire, 
 
518 THE ASSAY OP LEAD. 
 
 one inch long and forty centners in weight, had been placed 
 in a vertical position. The crucible charge, covered over 
 with two centners of decrepitated sodium chloride, was 
 smelted in a mineral coal muffle furnace, with the mouth 
 of the muffle closed, and the draught half open, at a mode- 
 rate temperature, the temperature then lowered for six to 
 seven minutes by opening the mouth of the muffle, then 
 the muffle closed again for an equal period, and the final 
 heat then given. The cessation of the low crackling of 
 the assay was now carefully attended to, and this, ceasing 
 after seven to eight minutes, indicated the completion 
 of the assay, The duration of the assay was twenty 
 minutes. 
 
 b. Roasting and Reduction Assay with Iron. One assay 
 centner of galena was roasted, at first at a low tempera- 
 ture, for about thirty minutes on a roasting dish, and the 
 dish then pushed into the back part of the muffle for six 
 to eight minutes to destroy the sulphates formed. The 
 roasted ore was rubbed fine, intimately mixed with three 
 hundred centners of black flux, and fifty centners of borax- 
 glass, placed in a crucible with a piece of iron at the 
 bottom, weighing twenty centners covered with salt, and 
 smelted as above. 
 
 c. Roasting and Fusing with Black Flux. One centner 
 of the roasted ore was smelted as before with three hun- 
 dred centners of black flux and fifty centners of borax, 
 but without iron. 
 
 The results obtained proved 
 
 1 . That with all those products which contain tolerably 
 pure lead sulphide, especially with high percentages, the 
 iron assay, a, gives in a remarkably predominant degree 
 the most lead (as high as 96 per cent, of all the lead 
 present). 
 
 2. With impure lead ores, which contain more foreign 
 sulphides, the assay a gives likewise the highest per- 
 centage, though the assays b and c give only a few per 
 cent. less. 
 
 3. If foreign sulphides are present in predominant 
 
ASSAY WITH SULPHURIC ACID. 519 
 
 quantity, the methods of b and c give a slightly higher 
 percentage than that of a. 
 
 LevoVs Fusion Assay with Potassium Ferrocyanide and 
 
 Cyanide. 
 
 According to Levol, the method of assaying galena for 
 its lead by smelting it with black flux and iron is defective 
 in two respects. First, it is difficult to choose precisely 
 the quantity of iron required for the reduction of the lead, 
 and a lack or excess of it either gives too little lead or a 
 button containing iron ; and second, in order that the 
 reaction may be complete and the lead unite to a button, 
 we are compelled to use a very high temperature, at which 
 lead volatilises. The first defect can indeed be removed 
 by the use of iron crucibles, but these are easily rendered 
 unserviceable, and require a pouring out of the fused 
 mass, and then globules of lead are apt to remain in the 
 slag. 
 
 By the use of a mixture of fifty parts of potassium 
 cyanide and one hundred of anhydrous potassium ferro- 
 cyanide to one hundred of galena, the loss of lead 
 diminishes to from 2 to 2^ per cent., probably in con- 
 sequence of the easy fusibility of the mixture, and the 
 extremely fine division of the iron in the potassium ferro- 
 cyanide. With antimonial galena this process is not 
 applicable, as the antimony is reduced and goes into the 
 lead. Potassium cyanide alone, gives, by reason of the 
 greater quantity of metallic sulphide which it retains, a 
 smaller product of lead. 
 
 6. ASSAY WITH SULPHURIC ACID. 
 
 The assay sample is rubbed as fine as possible. A 
 suitable quantity of it is then weighed out for an assay, 
 and boiled with four to eight times its weight of oil of vitriol 
 until all is decomposed. All excess of sulphuric acid is 
 then evaporated in a porcelain capsule, under a flue with a 
 good draught, and the mass carried to dryness. Boiling 
 
520 THE ASSAY OF LEAD. 
 
 sulphuric acid decomposes the sulphides, changing iron, 
 copper, nickel, zinc, &c., into salts which dissolve readily 
 in water, and also at the same time changing the lead 
 sulphide into sulphate, which in water, especially when 
 cold, and containing free sulphuric acid, is practically in- 
 soluble. The composition of the ore is in general ascer- 
 tained by first heating it with nitric acid or aqua regia, 
 and then with the addition of sulphuric acid, evaporating 
 to dryness. The dry mass, when cold, is moistened with a 
 small quantity of sulphuric acid, then cold water ; it is 
 afterwards, by the aid of a small brush, brought without 
 loss upon a small filter, and washed with cold water until 
 the filtrate is colourless. Unnecessary prolonging of the 
 washing is to be avoided, for lead sulphate is not absolutely 
 insoluble. The filter, with its contents, is dried in the 
 funnel, until it can be easily taken out of it without tear- 
 ing. It is now put immediately into the clay crucible in 
 which the lead sulphate is afterwards to be reduced, and 
 this is placed in a very gentle stove warmth. (Some 
 potassium carbonate may be first poured into the bottom 
 of the crucible.) When completely dry, the crucible with 
 the cover laid over it is very gently heated, so that the 
 filter carbonises, which very soon happens, as the free 
 sulphuric acid is not completely soaked out. The filter is 
 now stirred up with a little rod, black flux or potassium 
 carbonate with coal-dust and iron are introduced into the 
 crucible, and intimately mixed with the lead sulphate and 
 the rest of the insoluble residue. About four or five times 
 the volume of the whole residue is taken of black flux, 
 and the assay is further treated as prescribed in the portion 
 which follows upon the assaying of lead sulphate. 
 
 In this way the lead is concentrated, and the foreign 
 sulphides, which were specified above as the cause of the 
 failure of the assay in such cases, completely removed. 
 The result obtained in this way is satisfactory, and de- 
 serves the same confidence as one obtained in favourable 
 circumstances by the ordinary lead assay from an ore with 
 a medium or high percentage of lead. 
 
ASSAY OF GALENA. 52J 
 
 7. ASSAY OF GALENA IN THE WET WAY. 
 
 When in contact with metallic zinc, galena is readily 
 decomposed by acids. Even oxalic, acetic, and dilute 
 sulphuric acids are capable, when hot, of decomposing 
 galena metallic lead being deposited and sulphuretted 
 hydrogen gas set free, while with hydrochloric acid the 
 decomposition is peculiarly rapid and complete. 
 
 Galena is easily decomposed, also, even in the cold by 
 dilute nitric acid in presence of zinc ; but the reaction 
 differs in this case from that just described not metallic 
 lead but free sulphur is deposited, while lead nitrate goes 
 into solution. 
 
 The reaction with zinc and hydrochloric acid has been 
 employed with advantage by Mr. F. H. Storer, Professor of 
 Chemistry in the Massachusetts Institute of Technology, 
 for assaying galena, particularly the common American 
 variety, which contains no other heavy metal besides lead. 
 The details of the process are as follows : Weigh out 2 or 
 
 3 grammes or more of the finely powdered galena. Place 
 the powder in a tall beaker, together with a smooth lump 
 of pure metallic zinc. Pour upon the mixed mineral and 
 metal 100 or 150 c.c. of dilute hydrochloric acid which 
 has been previously warmed to 40 or 50 C. ; cover the 
 beaker with a watch-glass or broad funnel, and put it in a 
 moderately warm place. 
 
 Hydrochloric acid fit for the purpose may be prepared 
 by diluting 1 volume of the ordinary commercial acid with 
 
 4 volumes of water. For the quantity of galena above 
 indicated, the lumps of zinc should be about an inch in 
 diameter by a quarter of an inch thick ; they may be 
 readily obtained by dropping melted zinc upon a smooth 
 surface of wood or metal. 
 
 The zinc and acids should be allowed to act upon the 
 mineral during fifteen or twenty minutes in order to insure 
 complete decomposition. Any particles of galena which 
 may be thrown up against the cover or sides of the beaker 
 should of course be washed back into the liquid. It is 
 
522 THE ASSAY OF LEAD. 
 
 well, moreover, to stir the mixture from time to time with 
 a glass rod. 
 
 When all the galena has been decomposed, as may be 
 determined by the facts that the liquid has become clear, 
 and that no more sulphuretted hydrogen is evolved, decant 
 the liquid from the beaker into a tolerably large filter of 
 smooth paper, in which a small piece of metallic zinc has 
 been placed. Wash the lead and zinc in the beaker as 
 quickly as possible with hot water, by decantation, until 
 the liquid from the filter ceases to give an acid reaction 
 with litmus paper ; then transfer the lead from the beaker 
 to a weighed porcelain crucible. In order to remove any 
 portions of lead which adhere to the lump of zinc, the 
 latter may be rubbed gently with a glass rod, and after- 
 wards with the finger or a piece of caoutchouc, if need be. 
 Wash out the filter into an evaporating dish, remove the 
 fragment of zinc, and add the particles of lead thus collected 
 to the contents of the crucible. Finally, dry the lead at a 
 moderate heat in a current of ordinary illuminating gas, 
 and weigh. 
 
 The lead may be conveniently dried by placing the 
 crucible which contains it in a small cylindrical air-bath 
 of Eammelsberg's pattern, provided with inlet and outlet 
 tubes of glass, reaching almost to the bottom of the bath. 
 
 When the process is conducted as above described, 
 the lead undergoes no oxidation ; hence there is no occa- 
 sion for igniting the precipitate in a reducing gas. The 
 precipitate needs only to be dried out of contact with 
 the air. 
 
 If desirable, the sulphur in the galena can be esti- 
 mated at the same time as the lead, by arresting the 
 sulphuretted hydrogen in the ordinary way. 
 
 If the mineral to be analysed is contaminated with a 
 silicious or other insoluble gangue, the metallic lead may 
 be dissolved in dilute nitric acid after weighing, and the 
 insoluble impurity collected and weighed by itself. In the 
 case of galenas which contain silver, antimony, copper, or 
 other metals, precipitable by zinc, the proportion of each 
 metal must be estimated by assay or analysis in the usual 
 
ASSAY OF SUBSTANCES OF THE SECOND CLASS. 523 
 
 way, after the total weight of the precipitated metals has 
 been taken. 
 
 Besides galena, almost any of the ordinary lead com- 
 pounds may evidently be assayed by the method above 
 described. For example, metallic lead may be precipitated 
 quickly and completely from the sulphate, chromate, 
 nitrate, oxide, and carbonate and with peculiar ease from 
 the chloride by means of zinc and hydrochloric acid. The 
 method would also furnish an easy qualitative test for 
 the detection of baryta in white-lead. When applied 
 to the analysis of lead nitrate, it would probably be 
 best to decompose the nitrate by means of a solution of 
 sodium chloride before adding the zinc and hydrochloric 
 acid. 
 
 In all these cases the decomposition of the lead salt 
 by the zinc is so complete that no trace of coloration is 
 produced when sulphuretted hydrogen is added to the 
 liquid decanted from the metallic lead. 
 
 Attempts to estimate sulphur and lead in the same 
 portion of galena, by means of the reaction of zinc and 
 dilute nitric acid above described, have not given satisfac- 
 tory results. The free sulphur obtained by treating galena 
 with zinc and ordinary nitric acid, diluted with three, 
 four, and five volumes of water, always retains a small 
 quantity of lead, while a certain amount of sulphuric acid 
 is found in the clear liquid. It is, in short, well-nigh or 
 quite impossible to avoid the secondary reactions between 
 zinc and lead nitrate, and between sulphuric and nitric 
 acid, which set in as soon as, or just before, the last traces 
 of the galena have been decomposed. 
 
 CLASS II. 
 
 Assay of Substances of the Second Class. The assay of 
 these substances is very simple indeed. Litharge, minium, 
 lead carbonate, &c., may be assayed by simple fusion with 
 carbonaceous matter ; but when the operation is thus 
 conducted, loss of lead is sustained : it is therefore better 
 
524 THE ASSAY OF LEAD. 
 
 to add some flux which will readily fuse, and allow the 
 globules of reduced lead to collect into one button. No 
 flux fulfils this condition better than a mixture of sodium 
 carbonate and argol, which is to be intimately mixed with 
 the assay. The following is the best mode of procedure : 
 To 200 grains of the finely pulverised substance add 100 
 grains of argol, and 300 of sodium carbonate, and inti- 
 mately mix : place the mixture in a crucible which it 
 about half fills, and cover with a layer of common salt 
 about J inch thick ; submit the crucible to a very gradu- 
 ally increasing temperature, keeping the heat at low 
 redness for about a quarter of an hour ; then urging it to 
 bright red until the contents of the crucible flow freely ; 
 take it from the fire and shake, tap it as directed in the 
 copper assay, and either pour the contents into the mould 
 or allow to cool in the crucible. If the operator be 
 pressed for time, the mould may be used, but it is recom- 
 mended to allow the assay to cool in the crucible, for 
 unless the operator be very careful, and have had some 
 considerable practice, he is very liable to lose a small 
 .quantity of metal in the pouring. After the contents of 
 the mould or crucible, as the case may be, are cold, the 
 lead may be separated from the slag by repeated gentle 
 blows from the hammer : if any of the slag or crucible 
 adhere to the button, the latter may be readily freed from 
 it by placing the button between the finger and thumb 
 with its edge on the anvil, and then gently hammering it. 
 The lead will be so altered in shape under the hammer 
 that the slag or crucible readily falls off; and by continu- 
 ing the process the whole may be removed. The cleaned 
 button may then be hammered into a cubical form, and 
 is ready for weighing. 
 
 In the assay of lead great care must be taken in the 
 management of the temperature, as lead is sensibly volatile 
 above a bright red heat, even when covered with flux, 
 and still more so if any portion be uncovered from want 
 of sufficient quantity of flux ; neither must the assay 
 remain in after the flux flows freely, for a loss may thereby 
 occur from oxidation, by decomposition of sodium carbo- 
 
ASSAY OF SUBSTANCES OF THE SECOND CLASS. 525 
 
 nate, as explained in the reduction of copper ores and the 
 copper-refining process. 
 
 For the rationale of this mode of assay, refer to page 
 199, which explains the decomposition of lead oxide, with 
 the production of metallic ]ead, carbonic acid, and water, 
 by the agency of a substance, like argol, containing both 
 carbon and hydrogen. 
 
 Cupel bottoms, some lead fumes, and silicious slags 
 require a modified treatment in the assay, as the sub- 
 stances mixed with the oxide of lead (more particularly 
 bone-ash in the cupel bottoms) are very infusible ; and if 
 the flux already mentioned as applicable to the other 
 matters belonging to this class were employed, a very high 
 temperature would be necessary ; and as lead, as already 
 stated, is sensibly volatile above a bright red heat, an 
 evident loss of that metal would be the result. 
 
 Cupel bottoms may be thus assayed : 400 grains of the 
 finely pulverised bottoms to be mixed with 200 grains of 
 argol, 400 grains of sodium carbonate, and 400 grains of 
 pulverised fused borax ; the mixture is placed in a crucible 
 as already directed, covered with salt, and the fusion con- 
 ducted as just described. 
 
 Lead fumes and silicious slags require only half their 
 weight of fused borax, with 200 argol, 400 sodium car- 
 bonate, and 400 substance (fume or slag), covered with 
 salt. 
 
 The addition of the borax, which is a most powerful 
 flux, causes the fusion of the assay to take place almost as 
 readily with the last-named refractory substances as with 
 the former easily fusible and reducible matters. The 
 assay, however, is rather more subject to ebullition or 
 boiling over the sides of the crucible ; hence it must be 
 carefully watched ; and the instant it appears likely to do 
 so the crucible must be removed from the fire, gently 
 tapped on the furnace top, and when the effervescence 
 has subsided returned to the furnace, and this operation 
 repeated until the fusion proceeds tranquilly. 
 
 The lead obtained in these assays, if the ore or sub- 
 stance contained any foreign metal, is never pure ; if silver, 
 
526 THE ASSAY OF LEAD. 
 
 copper, tin, or antimony be present, the whole of either 
 of these metals will be found alloyed with the lead pro- 
 duced ; but if the ore contains zinc, and it be heated 
 sufficiently, only a trace remains ; nevertheless the zinc 
 carries off with it a considerable quantity of lead. 
 
 The following experiments will show what an influence 
 the presence of zinc has upon the return of lead : 
 
 100 parts of litharge, 
 100 parts of zinc oxide, 
 300 parts of black flux, 
 
 were fused together, and 84 parts of lead were the result. 
 100 parts of litharge, 
 100 parts of zinc oxide, 
 600 parts of black flux, 
 
 were fused together, and but 70 parts of lead were pro- 
 duced instead of 90, which the pure litharge ought to 
 have given. Hence it will be seen that the more zinc is 
 reduced, the more lead is volatilised. 
 
 If iron oxide be present in the assay, it is reduced, but 
 it remains in suspension in the slag, and the lead does not 
 contain a trace when it has not been too strongly heated. 
 If the assay be made at a very high temperature, the iron 
 may be fused, and then the lead will be ferruginous ; this 
 may be ascertained by means of the magnet. A similar 
 result was obtained by many assayers, who thought for a 
 long time that lead and iron could thus combine together ; 
 but by careful examination it is easily ascertained that the 
 ferruginous buttons are but mechanical mixtures of lead 
 and iron in grains. Indeed, by careful hammering, nearly 
 all the iron may be removed from the lead, so that it loses 
 its magnetic properties. 
 
 The manganese oxides, when mixed with the ore, are 
 changed into protoxide, which remains in the flux, and is 
 not reduced. 
 
 Wet Assay of Ores of the Second Class. Pulverise the 
 substance very finely, and to 100 grains placed in a flask 
 add one ounce of nitric acid diluted with two ounces 
 of water (if minium be the substance to be analysed, it 
 
ASSAY OF SUBSTANCES OF THE THIRD CLASS. 527 
 
 must be first heated to redness, so as to reduce the whole 
 of the lead it contains to the state of protoxide), and gently 
 heat, gradually raising the temperature to the boiling- 
 point : when all action seems to have ceased, pour the 
 contents of the flask into an evaporating basin, and evapo- 
 rate to dryness with the precautions directed in the analysis 
 of iron ore. Allow the dry mass to cool, add a little dilute 
 nitric acid, gently warm for an hour, then add water, boil 
 and filter. The whole of the lead now exists in the solu- 
 tion as nitrate : thus, say lead carbonate has been the 
 substance under analysis, then 
 
 PbO,C0 2 + N0 5 =PbO,N0 5 + C0 2 . 
 
 To the filtered solution containing the nitrate as above, 
 add solution of sodium sulphate, or dilute sulphuric acid, 
 until no further precipitation takes place ; insoluble sul- 
 phate of lead will now be thrown down : this must be 
 allowed to completely subside by standing in a warm place ; 
 and when the supernatant liquid is quite bright, the sul- 
 phate may be collected on a filter, washed, dried in the 
 water-bath, and weighed. It contains 68-28 per cent, of 
 metallic lead. 
 
 The decomposition of the lead nitrate by sodium sul- 
 phate may be thus expressed : 
 
 PbO,N0 5 + JSTa 2 0,S0 3 =PbO,S0 3 + Na 2 0,N0 5 . 
 
 Estimation of lead by standard solution will be 
 described at the end of this chapter. 
 
 CLASS III. 
 
 Assay of Substances of the Third Class. In the assay 
 of bodies belonging to this class a reducing agent must be 
 employed ; but if that alone be used, the sulphates and 
 arseniates produce sulphides and arsenides, and not pure 
 lead. The action of another reagent is therefore necessary, 
 in order to deprive the lead of the sulphur and arsenic 
 with which it is combined. 
 
528 THE ASSAY OF LEAD. 
 
 There are two reagents employed for the sulphates 
 the alkaline carbonates and metallic iron ; but for the 
 arseniates and arsenites iron must be employed, because 
 the alkaline carbonates have no action on the arsenides. 
 
 In all cases black flux is employed ; this furnishes a 
 reducing agent for the oxides, and a flux for the earthy 
 matters. Iron is added when the arsenites or arseniates 
 are assayed ; but that metal may either be employed or 
 not when the sulphates are operated upon. It is, how- 
 ever, always better to use it. 
 
 When a mixture of black flux and iron is employed, 
 the assay is made in exactly the same manner as that of 
 the sulphides (large nails are preferable whenever the use 
 of iron is indicated in a lead assay). With the sulphate, 
 the iron sulphide formed combines in the slag with the 
 alkaline sulphide ; but it is not so with the arseniates and 
 arsenites. The arsenide produced mixes neither with the 
 lead nor the slag, but gives rise to the formation of a brittle 
 matte which adheres slightly to the button of lead. 
 
 When only black flux is employed, either of the two 
 following processes may be adopted : First, the ore can 
 be fused with four parts of common black flux ; then, as 
 in the case of sulphides, the excess of carbon determines 
 the formation of a large quantity of an alkaline sulphide, 
 and consequently produces a desulphurisation of the lead. 
 Secondly, it may be fused with such a proportion of black 
 flux, containing only the requisite proportion of carbon to 
 reduce the lead oxide, or with an equivalent mixture of 
 sodium carbonate and charcoal. Pure lead sulphate fused 
 with one part of sodium carbonate and 4 per cent, of char- 
 coal gives 66 of lead ; but in order to employ this method 
 the richness of the ore must be known, and the dry way 
 is then useless, excepting for the estimation of the silver 
 these substances always contain. 
 
 Wet Assay of Substances of the Third Class. These 
 are treated in precisely the same manner as those of the 
 preceding class. 
 
 In treating lead ores with nitric acid a loss generally 
 results from the formation of insoluble lead sulphate. The 
 
ASSAY OF SUBSTANCES OF THE THIED CLASS. 529 
 
 solubility of these salts in sodium hyposulphite renders it 
 possible to avoid this inconvenience. After treatment with 
 nitric acid, J. Grceme (Bulletin de la Soc. Chimique de 
 Paris, November 5, 1873) proposes to exhaust the residue 
 with boiling water, until the soluble salts and the acid are 
 completely eliminated. It is then digested in the cold with 
 a concentrated solution of sodium hyposulphite. After this 
 treatment has been twice or thrice repeated the residue 
 is exhausted again with water ; the lead is then precipitated 
 from the filtrate by sulphuretted hydrogen or ammonium 
 sulphide ; to facilitate the agglomeration of the precipitate 
 and its washing, it is heated in the water-bath. The sul- 
 phide is then converted into sulphate, and its weight added 
 to that of the sulphate obtained directly. 
 
 F. Maxwell Lyte thinks the following process for the 
 management of the assay of lead in ores will be found 
 convenient, particularly where, as is often the case, the 
 lead to be estimated is mixed as sulphate with the matrix 
 insoluble in acid. 
 
 He dissolves the sulphate or chloride, as the case may 
 be, in ammonium acetate, makes the solution as neutral as 
 possible, and estimates the lead by a standard solution of 
 bichromate (a half-decinormal solution answers well) with 
 a silver nitrate indicator. 
 
 A. Mascazzini, previously to reducing the galena or 
 other lead ore to the metallic state, converts the lead 
 present in the ore into sulphate by igniting it in a porcelain 
 crucible with ammonium sulphate, after which the ore is 
 treated in the usual manner, The flux preferred by the 
 author is that recommended by Plattner, consisting of 
 13 parts of potassium carbonate, 10 of dry sodium car- 
 bonate, 5 of previously fused borax, and 5 of well-dried 
 starch. 
 
 To detect galena in mixtures, M. E. Jannesay throws 
 upon coarsely powdered galena a fragment of potassium 
 bisulphate, which gives a distinct evolution of sulphuretted 
 hydrogen. If the two bodies are ground , together the 
 odour becomes almost insupportable. Potassium bisul- 
 phate kept in fusion for half an hour produces the same 
 
 M M 
 
530 THE ASSAY OF LEAD. 
 
 effect, perhaps with less intensity. Sulphuric acid, mixed 
 or even heated with galena, does not give rise to a sen- 
 sible disengagement of sulphuretted hydrogen. Blende 
 gives a sulphydric odour, but less intense. Antimony, 
 iron, mercury, and silver sulphides give off no sensible 
 odour. Boulangerite, zinbrenite, and in general the sul- 
 phides in which lead and sulphur do not form an isolated 
 combination, do not evolve sulphuretted hydrogen with 
 the potassium bisulphate. 
 
 CLASS IV. 
 
 ALLOYS OF LEAD. 
 
 ASSAY WITH SULPHURIC ACID. 
 
 No docimastic assay is known for exhibiting the lead 
 isolated from its alloys. In individval cases a serviceable 
 result may be attained, if the metal with which the lead is 
 combined be estimated, and its quantity then deducted. 
 This method is, however, in general the more unreliable 
 the smaller is the quantity of lead, or when the lead is 
 alloyed with several metals ; so that then the quantity of 
 lead can often only be estimated by the partial or com- 
 plete aid of the wet way. 
 
 For many products (e.g. crude lead, hard lead con- 
 taining antimony or arsenic plumbiferous copper, &c.) 
 the assay with sulphuric acid described on page 519 is 
 suitable. One assay centner of the substance is decom- 
 posed by nitric acid or aqua regia, then, with the addition 
 of sulphuric acid, evaporated to dryness, and the dry 
 mass treated as above directed. If the residue consists 
 only of lead sulphate, it can be brought upon a weighed 
 filter, and from the weight of the residue after drying, 
 the amount of lead may be calculated. 100 parts lead 
 sulphate contain 68-33 parts lead. 
 
DOCIMASTIC ESTIMATION OF LEAD. 531 
 
 ESTIMATION OF LEAD BY MEANS OF STANDARD SOLUTIONS. 
 1. FLORES DUMONTE'S METHOD. 
 
 Estimation of Lead by means of Standard Solutions. 
 This process is due to M. Flores Dumonte, and may be 
 thus described : This mode of analysis is analogous to 
 that proposed by Pelouze for the estimation of copper ; 
 advantage is taken of the fact that oxide of lead is soluble 
 in caustic potash in the same manner that copper oxide is 
 soluble in ammonia ; and from either solution the respective 
 metal is precipitated by means of a standard solution of 
 sodium sulphide. 
 
 The solution of sodium sulphide may be conveniently 
 made by dissolving one ounce of sodium sulphide in one 
 quart of water, and estimating how much of it is necessary 
 to precipitate twenty grains of lead. To this end weigh 
 off twenty grains of lead, dissolve them in nitric acid, di- 
 lute with water, and add excess of caustic potash until the 
 oxide of lead first thrown down is completely dissolved. 
 The solution must now be heated to ebullition, and the 
 sodium sulphide gradually added from the burette ; at 
 each addition a black precipitate of lead sulphide falls. 
 The liquid is then boiled for a short time, by which means . 
 it brightens ; more sodium sulphide is then added, and the 
 whole again boiled, and these operations alternately con- 
 tinued until no further coloration or blackening is pro- 
 duced by the last drop of sulphide. The number of 
 divisions used is then read off, and the calculation made 
 as in Pelouze's copper assay, substituting lead for copper. 
 
 Having thus standardised the solution of sodium sul- 
 phide, the assay of a sample of ore may be thus made : 
 If the ore belong to the first class, dissolve it in dilute 
 nitric acid and evaporate to dryness ; to the dry mass add 
 excess of caustic potash solution, and boil ; after about 
 a quarter of an hour's ebullition, filter and throw down 
 the lead as directed, with the standard solution ; from the 
 amount used calculate the quantity of lead present ; if the 
 ore be of the second or third class, treat with strong nitric 
 
 M M 2 
 
53-2 THE ASSAY OF LEAD. 
 
 acid and sodium carbonate, as already directed. The lead 
 carbonate so produced may be dissolved in either nitric or 
 acetic acid, and to the solution thus obtained add caustic 
 potash, &c. 
 
 2. SCHWARTZ'S METHOD. 
 
 Dissolve 14*730 grammes of pure potassium bichromate 
 in sufficient water to form one litre. One cubic centimetre 
 of this solution precipitates 0-0207 gramme of lead. 
 
 In the estimation of pure lead a certain quantity of 
 it should be dissolved in a minimum of nitric acid, the 
 solution diluted with water, carefully neutralised with 
 ammonia or sodium carbonate, excess of sodium acetate 
 added, and the solution precipitated by the potassium 
 bichromate solution. When the precipitation approaches 
 its end, or when the precipitate commences readily to 
 subside, some drops of a neutral solution of silver nitrate 
 are deposited on a porcelain plate, and the potassium 
 chromate solution only added by two or three drops at a 
 time to the liquid under examination ; after each addition 
 the whole is well stirred, allowed to subside, and a drop of 
 the clear supernatant liquor added to one of the drops of 
 the silver solution. As soon as the potassium bichromate 
 is in excess the two drops form a red colour, while the 
 precipitated lead chromate has no effect on the silver test, 
 but simply floats on the top as a yellow precipitate. 
 Should the solution assume a yellow colour before the 
 silver reaction has commenced, it would indicate that not 
 sufficient sodium acetate had been added in the first in- 
 stance, and it w r ould be necessary to add this now, and 
 also a cubic centimetre of a normal lead solution, contain- 
 ing 0-0207 of lead as nitrate. The slight turbidity which 
 first takes place soon goes off, and the operation may be 
 proceeded with as before. One cubic centimetre must 
 naturally, in such instance, be deducted from the amount 
 of chrome solution, on account of the extra addition of 
 lead. 
 
 Bismuth alone seems to interfere with the reaction, 
 
BUISSON S VOLUMETRIC PROCESS. 533 
 
 .and behaves very like lead with chromic acid, and if present 
 it requires a different mode of proceeding. 
 
 The higher oxide of mercury is not precipitated by 
 potassium bichromate, not even in an acetic solution, while 
 the lower oxide is ; and, as it is difficult to peroxidise all 
 the mercury when united with lead, even by long-continued 
 boiling in nitric acid, it becomes necessary to evaporate and 
 calcine the residue till all the mercury is volatilised. To 
 obviate the formation of red-lead, the calcined residue has 
 to be moistened with a few drops of oxalic acid, and again 
 dried and carefully calcined and dissolved in acetic acid ; 
 after this the lead may be estimated as usual. To avoid 
 the above calcinations, the mercury may be precipitated 
 from the nitric acid solution by means of hydrochloric acid, 
 and the liquid boiled till the calomel is converted into the 
 higher chloride. 
 
 Copper, cadmium, zinc, iron, and cobalt do not in the 
 least interfere with the reaction, provided the iron is per- 
 oxidised. Of the different acids, hydrochloric acid some- 
 what disturbs the last silver reaction, but by using larger 
 drops, and allowing the reaction of silver chloride to go 
 off, we obtain the usual silver chromate reaction. 
 
 Lead sulphate has first to be converted into the state of 
 carbonate, by boiling with sodium carbonate, when it may 
 be dissolved in acetic acid. Lead phosphate and arsenite, 
 -or other lead salts insoluble in acetic acid, may be dis- 
 solved in nitric acid, and estimated according to the older 
 .method. 
 
 .< 
 o. BUISSON'S VOLUMETRIC PROCESS FOR ESTIMATING LEAD. 
 
 This process is based on the precipitation of lead by 
 potassium bichromate and the decomposition of the excess 
 of the reagent used by potassium iodide in a liquid 
 acidulated with sulphuric acid. 0'5 to 1 gramme of the 
 pulverised mineral to be tested is dissolved in dilute nitric 
 .acid. The solution is saturated with potash, and the preci- 
 pitate redissolved in acetic acid. The iron is removed by 
 boiling. To the solution separated from the iron 25 c.c. 
 
534 THE ASSAY OF LEAD. 
 
 of potassium bichromate is added, and water to bring the 
 volume up to 250 c.c. After standing for some time the 
 solution is filtered through a dry filter-paper. To 100 c.c. 
 of .the clear liquid an excess of dilute sulphuric acid and 
 potassium iodide are added in such a way as to redissolve 
 the iodine set free, and then a few cubic centimetres of 
 solutions starch. By means of a graduated burette, sodium 
 hyposulphite is added until decolourisation of the starch 
 iodide takes place. The difference of standard obtained by 
 treating the bichromate alone, and after precipitation with 
 a known weight of lead, gives a basis for calculating the 
 amount of metal contained in the substance tested. 
 
 Silver, bismuth, copper, and baryta should be separated 
 from the lead before applying this process. 
 
 4. W. DIEHL'S PKOCESS. 
 
 W. Diehl proposes the following process for the 
 volumetrical estimation of lead ( 4 Zeitschrift fur Analy- 
 tische Chemie '). He employs a ^ normal solution of 
 potassium bichromate, containing 7 '38 grnis. per litre, 
 each c.c. representing 0*01035 grin, of lead, and a solution 
 of sodium hyposulphite, containing 4 to 5 grms. per litre. 
 To determine the relation between these two solutions, 
 20 to 30 c.c. water are mixed with 20 to 25 dilute sul- 
 phuric acid (1 vol. monohydrated acid, and 2 vols. of 
 water) ; a certain excess of sulphuric acid is indispensable, 
 hydrochloric acid being less convenient. The liquid is 
 brought to a boil, and the solution of hyposulphite is 
 added drop by drop. The solution becomes gradually 
 paler in colour. Towards the end, after the addition of a 
 few drops, it is let boil up again. The end of the reaction 
 may generally be distinguished by the liquid turning per- 
 fectly colourless, a result occasioned by a single drop. In 
 order to judge of the colour, the flask towards the end of 
 the operation may be set in a porcelain capsule. When 
 very large quantities of bichromate are used the liquid 
 does not become perfectly colourless, but slightly greenish. 
 
 In assaying ores in this manner they are dissolved 
 
DIEHL'S PKOCESS. 535 
 
 in aqua regia and dilute sulphuric acid, the solution con- 
 centrated till the sulphuric acid begins to evaporate, 
 diluted with water, boiled to dissolve ferric sulphate, let 
 cool, and filtered through a smooth filter, washing with 
 water containing sulphuric acid. To the residue in the 
 flask as little as possible of which is thrown upon the 
 filter is added about 15 c.c. of a solution of neutral am- 
 monium acetate, and about 50 c.c. of water. The whole is 
 then boiled, and filtered through the same filter, into which 
 a drop of ammonia has been put, into a flask. The same 
 operation is then repeated with 5 c.c. ammonium acetate, 
 and the residue is finally well washed with boiling water, 
 to which a little of the same salt has been added. Thorough 
 washing is necessary, since filters have been found to 
 retain ammonium and lead acetate and tartrate with con- 
 siderable obstinacy. It is then advisable further to wash 
 the filter from its margin downwards with a little boiling 
 dilute hydrochloric acid (1 part hydrochloric acid at sp. 
 gr. 1-12 with 10 parts of water), and then to wash again 
 with hot water. In this manner every trace of lead is re- 
 moved from the filter. A thin filter-paper should be used, 
 and should be washed previously. The solution of lead 
 sulphate in ammonium tartrate thus obtained is then 
 titrated in the cold with potassium bichromate ; with the 
 aid of heat, ammonium acetate dissolves a little lead 
 chromate. The precipitate settles readily if the flask is 
 shaken, and the end of the reaction can be observed to 
 within 0-2 to 0*4 c.c. 
 
 An excess of at least 2 c.c. of the chromate solution 
 should be added, in order to obviate the solubility of the 
 lead salt. It is advisable in every experiment to take as 
 closely as possible an equal quantity. After thoroughly 
 shaking, it is allowed to stand for half an hour and filtered. 
 If the liquid passes through turbid, a few drops of a 
 solution of sodium acetate, acidulated with acetic acid, are 
 added. If, after all, a little lead chromate passes through 
 the filter, the filtration is repeated. The precipitate is 
 washed four times with cold water, and the solution is 
 acidulated with sulphuric acid and titrated as above. 
 
636 THE ASSSAY OF LEAD. 
 
 Ammonium acetate is preferable to all other ammonium 
 salts as a solvent for lead sulphate. It is applied in a 
 neutral or faintly acid state. Free ammonia renders the 
 solution turbid. 1 grm. lead sulphate requires 15 c.c. 
 of the liquid acetate for solution. Ammonium tartrate 
 cannot be used. 
 
537 
 
 GHAPTEE XII. 
 
 THE ASSAY OF TIN. 
 
 THIS metal is always found by the assayer in the state of 
 oxide. 
 
 Tin Oxide (Sn0 2 ). The appearance of this mineral 
 gives no indication, excepting to an experienced eye, that 
 metallic matter enters largely into its composition ; yet its 
 great density would lead one to suppose such to be the case. 
 Its colour varies from limpid yellowish white to brownish 
 black and opaque, passing from one to the other by all 
 intermediate shades. It usually possesses a peculiar kind of 
 lustre which cannot be readily described, but once seen can 
 scarcely be mistaken. It occurs crystallised in square prisms, 
 terminated by more or less complicated pyramids. These 
 crystals, derived from the octahedron, are often macled or 
 hemi tropic, so that they often possess re-entering angles, 
 which is to a certain extent characteristic. The principal 
 varieties are the following : 
 
 1. Crystallised Tin Oxide is found in more or less 
 voluminous crystals of the colour and form as above. 
 
 2. Disseminated Tin Oxide. This variety occurs in 
 grains of various sizes, sometimes so small as not to be 
 visible to the naked eye. It is found in the primitive 
 rocks. 
 
 3. Sandy Tin Oxide forms pulverulent masses often 
 of great extent ; in appearance it is merely a brown 
 sand. 
 
 4. Concretionary Tin Oxide, Wood Tin. This variety 
 occurs in small mamellated masses, the fibrous texture of 
 which resembles that of wood ; hence the name. 
 
538 THE ASSAY OF TIN. 
 
 The following is an analysis of a sample of tin oxide 
 from Cornwall : 
 
 Tin 77-50 
 
 Oxygen 21-40 
 
 Iron -25 
 
 Silica . -75 
 
 The following remarks on tin ore and the minerals 
 which may be mistaken for it are from the pen of Dr. A. 
 Leibius, Senior Assayer of the Sydney branch of the Eoyal 
 Mint. 
 
 The colour of native tin ore varies from white to pink, 
 ruby-red, grey, greyish-black to black ; it therefore is cer- 
 tainly no very reliable criterion for distinguishing tin ore. 
 
 A safer characteristic is the weight, or specific gravity, 
 that of tin ore being 6-8 to 7*0. Unfortunately, however, 
 the specific gravity of iron tungstate is nearly the same 
 as that of tin oxide in fact, a little higher, being 7*19 to 
 7-55. 
 
 Titaniferou.s iron has a specific gravity of from 4 f 5 to 
 5-0, and magnetic iron 4-9 to 5*2. 
 
 Basaltic hornblende and iron silicate have also been 
 mistaken for tin ore ; but the specific gravity of the former 
 being only 3-1 to 3-4, and that of the latter 3-8 to 4*2, 
 ought to have prevented such mistakes. The colour of the 
 powdered ore forms a much better criterion than that of 
 the unpowdered ore. The powder of good tin ore varies 
 only from whitish-grey to dark drab, while iron tungstate 
 powders reddish-brown, and titaniferous iron black. 
 
 Most of the minerals appear to be mistaken for tin ore 
 on account of their dark granular pieces having been taken 
 for such ; but Mr. Leibius has seen a sample which con- 
 sists of blackish pieces with about 50 per cent, of small 
 indistinct crystals of a pink and dark ruby colour, with a 
 few small white crystals. The whole mixture being pretty 
 heavy has certainly at first sight all the appearance of 
 good tin ore. Even on closer inspection, when the darker 
 portion just referred to might have been recognised as an 
 iron compound, the ruby-coloured portion might readily 
 pass muster for tin ore unless chemically examined. On 
 further examination the whole sample was found to be 
 
ASSAY OF PURE OXIDE OF TIN. 539> 
 
 free from tin. It was found to consist 1. Black portion, 
 about 50 per cent, of the sample, having a specific gravity 
 of 4-47, was found to be titaniferous iron. 2. Buby-coloured 
 and dark red pieces, about 50 per cent, of sample, with a 
 specific gravity of 4*57, were found to be zircons or hya- 
 cinths, showing the characteristic property, mentioned 
 in Professor Thompson's excellent ' Guide to Mineral Ex- 
 plorers ' (see ante, pp. 254,293), of becoming completely 
 and lastingly colourless when exposed to heat before the 
 blowpipe. 3. Besides these zircons were found a few small 
 topazes and garnets, and also a small sapphire. The 
 specific gravity of the mixed sample as received was 
 4-55. 
 
 As already mentioned, there was actually no tin in the 
 sample, and it forcibly illustrates the necessity of the 
 precaution, in dealing with tin ore, to have it carefully 
 assayed. 
 
 Assay of Pure Tin Oxide. Pure tin oxide may be 
 very readily assayed in the following manner : Weigh off 
 400 grains, place them in either a black-lead or charcoal- 
 lined crucible, cement on a cover by means of Stourbridge 
 clay, and place in the fire. The heat should for the first 
 quarter of an hour be a dull red, after which it may be 
 raised to a full bright red for ten minutes, and the crucible 
 removed with care so as not to agitate or disturb the con- 
 tents ; tapping in this case must not be resorted to. When 
 the crucible is cold, remove the cover, and a button of 
 pure tin will result ; this weighed and divided by four 
 gives the percentage. If the operation has not. been care- 
 fully conducted it sometimes happens the tin is not in one 
 button, but disseminated in globules either on the charcoal 
 lining or on the sides of the black-lead pot ; in this case 
 the charcoal on the one hand, or the black-lead crucible 
 on the other, must be pulverised in the mortar and passed 
 through a sieve ; the flattened particles of tin will be re- 
 tained by the sieve, and can be collected and weighed. If 
 any small particles escape the sieve, they may be separated 
 from the lining or crucible by vanning. 
 
 If a charcoal or black-lead crucible be not at hand, an 
 
-540 THE ASSAY OF TIN. 
 
 ordinary clay pot may be used, but not so successfully, 
 excepting under certain circumstances to be hereafter 
 described. Indeed, in Cornwall the ordinary mode of con- 
 ducting this assay is in a naked crucible, thus : About 2 
 ounces of the ore are mixed with a small quantity ol 
 culm, and projected into a red-hot crucible. If the ore 
 seems to fuse or work sluggishly, a little fluor-spar is 
 added, and after about a quarter of an hour's fusing at a 
 good high temperature the reduced and fused tin is poured 
 into a small ingot mould, and the slag examined for metal 
 by pounding and vanning. This method never gives the 
 whole of the metal. To effect this, without fear of mis- 
 chance in the assay sometimes occurring, as already de- 
 scribed with both black-lead and charcoal-lined crucibles, 
 it may be thus conducted ; always supposing the oxide 
 to be pure, or nearly so, or at least containing little 'or no 
 silicious matter. 
 
 To 400 grains of ore add 100 grains of argol, 300 grains 
 of sodium carbonate, and 50 grains of lime ; mix well 
 together, place in a crucible which the mixture half fills, 
 cover with a small quantity of sodium carbonate and 200 
 grains of borax. Place the whole in the furnace with the 
 necessary precautions, raise the heat very gently, and keep 
 it at or below a dull red heat for at least twenty minutes ; 
 then gradually increase until the whole flows freely. Ee- 
 move the crucible, tap it as for copper assay, and allow to 
 cool. When cold, break it, and a button of pure metallic 
 tin will be found at the bottom, and a flux perfectly free 
 from globules and containing no tin. 
 
 There is. yet another process, which is more easy of 
 execution ; but the reagent employed is more expensive, 
 not so readily obtainable, and more difficult to keep without 
 decomposing than any of the substances above employed. 
 This reagent has been mentioned, in another part of this 
 volume, as a blowpipe flux, and, in the assay of copper 
 ores by standard solutions, as potassium cyanide. This is 
 the most effective reducing flux for tin ores yet known. It 
 acts by absorbing oxygen to form a compound known 
 as potassium cyanate. 
 
ASSAY OF PURE OXIDE OF TIN. 541 
 
 The assay, by means of this substance, may be made 
 in ten minutes. 
 
 This method of estimating the value of tinstone has 
 been frequently practised by the writer, and has uniformly 
 furnished correct results with but little expenditure of 
 time and labour. The method of operating is as fol- 
 lows : The sample, having been carefully selected, is first 
 crushed by the hammer in a steel mortar, and then 
 further reduced to powder in an agate mortar. 100 grains 
 is a convenient quantity to be taken for analysis, and it 
 is always advisable to make two independent experiments 
 upon the same sample of ore, with the view of having a 
 control, and the highest result obtained is that upon which 
 to place reliance, since the error must always be on the 
 side of loss rather than excess. A couple of small Hessian 
 crucibles, of about 3 oz. capacity, are prepared in the first 
 instance by ramming into the bottom of them a small 
 charge of powdered potassium cyanide sufficient to form 
 a layer of about half an inch in depth ; the weighed quan- 
 tities of tin ore are then intimately mixed with from four 
 to five times their weight of the powdered cyanide, and 
 the mortar rinsed with a smull quantity of the pure flux, 
 which is laid upon the top of the mixture. The crucibles 
 are then heated in a moderate fire, or over a gas blowpipe, 
 and kept for the space of ten minutes at a steady fusion ; 
 they are then removed, gently tapped to facilitate the 
 formation of a single button, and allowed to cool. Upon 
 breaking the crucibles the reduced metal should present 
 an almost silvery lustre, with a clean upper layer of melted 
 flux. It is advisable to dissolve the latter in water, in order 
 to be certain as to the absence of any trace of reduced metal 
 or heavy particles of the original ore. There is always 
 contained in the commercial cyanide a sufficient quantity 
 of alkaline carbonate to secure the perfect fusion of the 
 silicious gangue, and other like impurities in the tin ore, 
 but the operator should assure himself of the absence of 
 copper and lead in the ore, either by preliminary treat- 
 ment with hydrochloric acid, in which tinstone is ab- 
 solutely insoluble, or by testing the button of reduced tin 
 
542 THE ASSAY OF TIN. 
 
 after hammering or rolling for such metallic admixture. 
 We have usually found a minute trace of iron, and some- 
 times gold in the melted buttons, but not so much as to 
 add appreciably to their weight. 
 
 When worked with ordinary care, this process may be 
 relied upon as giving numbers true to within -J per cent., 
 and we do not know any other method which exceeds this 
 in accuracy and rapidity of execution. The following are 
 a few analytical results taken at random from a number 
 of ores assayed in this manner : 
 
 Tin per cent. 
 L ~~n. 
 
 Sample No. 1 . . . . 45-6 45-8 
 
 No. 2 .... 57-2 57-6 
 
 No. 3 . . . . 68-4 68-7 
 
 Assay of Tin Oxide mixed with Silica. Although tin 
 oxide is completely reducible by charcoal or other carbo- 
 naceous matter, yet it has such an affinity for silica, that 
 whenever that substance is present the metal cannot be 
 wholly reduced, excepting at the highest temperature of a 
 wind furnace. The following experiments will show the 
 influence of silica on the return of tin in an assay of oxide 
 of that metal with black flux. 
 
 Ore . . 100 100 100 100 100 
 Quartz . 25 66 100 150 300 
 
 The first gave 52 per cent, of tin ; the second, 43 per 
 cent. ; the third, 28 per cent. ; the fourth, 10 per cent. ; 
 and the last, nothing. 
 
 The slags also produced in the treatment of tin ores in 
 the large way give no return with black flux. This mode 
 of assay, however, has been recommended by some, but 
 from the foregoing experiments is proved to be perfectly 
 fallacious : that is, unless the quantity of silica present be 
 very small in comparison to the amount of tin oxide ; and 
 even when the latter is present in four times the quantity 
 of the silica, as in experiment No. 1, a loss of 20 per cent, 
 of tin is sustained. 
 
 Assay of Tin Ores containing Silica and Tin Slags. It 
 having just been shown how injuriously the presence of 
 silica influences the produce of tin, both in ores and slags, 
 
ASSAY OF TIN ORES CONTAINING SILICA. 543 
 
 other methods of assay than those just described must be 
 adopted for such substances. These will now be detailed. 
 
 Tin ores containing silica may be treated by two 
 methods : in the first the silica must be carefully separated 
 by vanning ; if the ore be well pulverised this is the best 
 and most expeditious method. In conducting this assay 
 take 400 or more grains of the pulverised ore, according 
 to its richness (if poor, as much as 2,000 grains may be 
 taken), van it carefully, dry the enriched product, which 
 will, if the operation has been properly conducted, be 
 nearly pure oxide of tin, and assay it as already described 
 for ores containing no silica. The other process of assay 
 may be thus conducted, and is dependent upon the fact that 
 iron displaces tin from its combination with silica : thus, 
 if a compound of tin oxide and silica be heated to white- 
 ness with metallic iron, a portion of the iron oxidises and 
 replaces the tin oxide, which was previously in combina- 
 tion with the silica as a tin silicate, and metallic tin and 
 iron silicate result, the tin so reduced combining with any 
 metallic iron that may be in excess, and the button thus 
 obtained is an alloy of tin and iron, whilst the slag is 
 entirely deprived of tin. 
 
 In this kind of assay mix 400 grains of the silicious 
 tin oxide with 200 grains of iron oxide (either pulverised 
 hematite or forge-scales will answer this purpose), 100 
 grains of pounded fluor-spar, and 100 grains of charcoal 
 powder : place the mixture in a crucible and cover with 
 a lid, gradually heat to dull redness, and keep at that 
 temperature for half an hour, then heat to whiteness for 
 another half-hour, and remove the crucible from the fur- 
 nace, allow to cool, and break. The button so obtained is 
 to be treated in the wet way, as hereafter described. 
 
 The assay of tin slags is conducted in the same manner, 
 or simply by mixing the pulverised slag with 20 per cent, 
 of iron filings, and fusing. 
 
 Assay of Tin Ores containing Arsenic, Sulphur -, and 
 Tungsten (Wolfram). In the assay of such ores it is neces- 
 sary to remove arsenic, sulphur, and tungsten before 
 attempting to obtain the tin in a pure state by the dry 
 
544 THE ASSAY OF TIN. 
 
 assay. Ores of tin which contain either one or all of 
 these substances are most common ; hence this mode of 
 treatment will be generally required. 
 
 Most assayers usually submit the ore to the same mode 
 of treatment which it undergoes on the large scale by cal- 
 cination, or rather roasting, by which the greater part of 
 the arsenical and pyritic matter is removed ; this process 
 fails, however, to remove the whole of these substances, 
 and does not at all affect the tungsten. The following 
 process is therefore preferable, and is founded on the fact 
 that arsenical and other pyrites, as well as iron tungstate 
 (wolfram usually accompanying tin ores), are completely 
 decomposed by nitro-hydrochloric acid (aqua regia) at the 
 boiling temperature, the oxide of tin alone not being 
 affected : Take 400 grains or more of the impure tin 
 sample, place them in a flask, and add 1^ ounce of hydro- 
 chloric acid, and -J an ounce of nitric acid, heat gently for 
 about half an hour, and then boil until the greater part of 
 the mixed acids has evaporated ; the sulphur and arsenic 
 will by this time be converted into sulphuric and arsenic 
 acid, and the wolfram completely decomposed, its iron and 
 manganese having become soluble, and its tungstic acid 
 remaining in the insoluble state with the oxide of tin and 
 any silica that may be present. Allow the flasks and con- 
 tents to cool, add water, allow to settle, and decant, and so 
 on until the water passes off tasteless. The insoluble matter 
 in the flask is now tin oxide, silica, and tungstic acid ; to 
 remove the latter, digest for an hour at a very gentle heat 
 with one ounce of solution of caustic ammonia, with occa- 
 sional agitation ; add water, and van the remainder to 
 separate silica ; nothing remains now but tin oxide, with 
 perhaps a little silica : this is now to be dried and assayed 
 as directed for ores containing little or no silica. 
 
 If only an approximative assay be needed, it may be 
 accomplished after this treatment by taking the specific 
 gravity of the remaining oxide ; so that all ores of tin may 
 be thus roughly assayed, it being premised that the above 
 operation has been so carefully performed that nothing 
 but tin oxide and silica remain. The specific gravity of 
 
ASSAY OF TIN OEES CONTAINING AESENIC, ETC. 
 
 545 
 
 the thus purified ore is to be taken. All now that is neces- 
 sary to be known is the specific gravity of tin oxide, its 
 percentage of pure tin, and the specific gravity of silica, 
 and a simple calculation gives the result. The following 
 is the formula : 
 
 Let a represent the specific gravity of tin oxide. 
 b silica. 
 
 c the mixture left after treatment 
 
 with acid, &c. 
 w weight of rough tin oxide or mixture left after 
 
 treatment with acid, &c. 
 x tin oxide. 
 
 y silica. 
 
 And y = 
 
 c (a b) 
 
 Or in arithmetical form thus : 
 
 1. From the specific gravity of the rough tin oxide (mixture of tin oxide 
 and silica) deduct the specific gravity of the silica. 
 
 2. Multiply the remainder by the specific gravity of the tin oxide. 
 
 3. Multiply the weight of the rough tin oxide by the last product, which 
 will make a second product, which may be called P. 
 
 4. From the specific gravity of tin oxide deduct the specific gravity of 
 silica. 
 
 5. Multiply the difference by the specific gravity of the rough tin oxide. 
 
 6. Take this product for a divisor to divide the above product P : the quo- 
 tient will be the weight of pure tin oxide in the rough oxide, and the quantity 
 of metal can now be readily calculated. 
 
 The following is an assay worked out in this manner : 
 
 400 grains of the ore are treated with nitro-hydrochloric acid and ammonia 
 as above described, washed, and dried. Suppose the dried matter weighs 
 250 grains. The 250 grains thus obtained are placed in the specific gravity 
 bottle, and the specific gravity is found to be 5*4. 
 
 Specific gravity of tin oxide (approximate) . * . 6*9 
 silica' . . 2*6 
 
 Sp. Gr. 
 Bough Oxide 
 
 5-4 
 
 2-8 
 
 Weight of 
 Bough Oxide 
 
 250 
 
 Sp. Gr. 
 Pure Oxide 
 
 6-9 
 
 4-3 
 
 Sp. Gr. 
 
 Silica 
 
 2-6 
 
 Sp. Gr. 
 Pure Oxide 
 
 6-9 
 
 19-32 
 
 Sp. Gr. 
 Silica 
 
 2-6 
 
 Sp. Gr. 
 Bough Oxide 
 
 5-4 
 
 4830 =208-4 
 
 2-8 
 
 19-32 
 4830 
 
 4-3 
 23-22 
 
 23-22 
 
 N N 
 
546 THE ASSAY OF TIN. 
 
 208*4 grains is therefore the weight of pure oxide in the 400 grains of ore. 
 Now tin oxide contains 78-61 parts of pure tin, and 
 208-4 x 78-61 
 
 100 
 
 So that 400 grains of rough tin ore contain 163*72 grains of pure tin, and 
 
 =40-93. 
 
 The rough sample first operated on contains, therefore, 
 40*93 per cent, of metallic tin. 
 
 Mr. J. H. Talbott's Process for assaying Tin in the Pre- 
 sence of Tungsten. Another satisfactory method of assaying 
 tin in the presence of tungsten has been described by Mr. J. 
 H. Talbott. 
 
 The method is based on the fact already mentioned 
 that tin oxide is reduced by potassium cyanide with great 
 facility ; while tungstic acid undergoes no reduction, even 
 when heated with the cyanide at a high temperature. The 
 tin and tungsten oxides are to be heated in a porcelain 
 crucible with three or four times their weight of commer- 
 cial potassium cyanide previously fused, pulverised, and 
 thoroughly mixed with the two oxides. The mass is kept 
 fused for a short time, when the tin separates in the form 
 of metallic globules, while the tungstic acid unites with 
 the alkali of the potassium cyanate and carbonate present. 
 After cooling, the mass is to be treated with hot water, 
 which dissolves the alkaline tungstate and other salts, and 
 leaves the tin as metal ; this, is to be separated by filtra- 
 tion, washed, dried, and weighed as tin oxide, after oxida- 
 tion in the crucible with nitric acid. The tungstic acid 
 may be estimated by difference, or be precipitated by 
 mercury protonitrate, after boiling the solution with nitric 
 acid to decompose the excess of potassium cyanide present, 
 and then re-dissolving the precipitated tungstic acid by 
 means of an alkali. 
 
 Estimation of Tin by the Wet Method. There are 
 several methods of effecting this analysis, the chief diffi- 
 culty- being found in the intractable nature of the tin oxide, 
 it resisting the action of all acids. This, however, may be 
 overcome as first shown by Klaproth, who found that very 
 finely levigated tin oxide was soluble in hydrochloric acid 
 
ESTIMATION OF TIN IN THE WET WAY. 547 
 
 after a prolonged fusion with caustic potash. The following 
 is his process : 
 
 Fifty grains of the tin ore, reduced to the most minute 
 state of division by levigation or otherwise, are mixed with 
 four times its weight of caustic potash. The best mode of 
 mixing is to place the caustic potash in a silver crucible, 
 add its own weight of water, and apply a gentle heat until 
 the potash is dissolved ; then stir in tin ore, and gradually 
 evaporate to dryness, stirring all the time to prevent loss by 
 spirting, as in the analysis of ironstone : when thoroughly 
 dry, enclose the silver crucible in one of clay, and submit 
 the whole to a dull red heat for at least half an hour ; 
 rather more than less renders the perfect solution of the 
 tin oxide more certain. When cold, act on the contents 
 of the crucible with dilute hydrochloric acid, transfer the 
 liquid and any undissolved matter to a flask, add some 
 strong hydrochloric acid, and boil for half an hour. If at 
 the end of this time any of the tin ore remains unacted on, 
 it must be separated by decantation or otherwise from the 
 solution, dried, again fused with potash, and then treated 
 with hydrochloric acid, in which it will now be found 
 totally soluble. This second operation will not be needed 
 if care has been taken to reduce the ore to the finest 
 possible state of division at first. The solution, however 
 obtained, is to be evaporated to dryness, and when cold 
 treated with a small quantity of hydrochloric acid, allowed 
 to stand for half an hour, then water added, boiled, and 
 filtered : the whole of the tin will pass through in solution 
 as tin chloride, and any silica or tungstic acid that may 
 be present will remain in the filter. If the or$ contains 
 copper, lead, and iron, these metals will also be in solution 
 at all events, the lead partially so ; but if the ore has, 
 previously to its fusion with caustic potash, been treated 
 with aqua regia, as already described, then it will contain 
 tin alone. It is always better thus to separate foreign 
 matters before attempting the solution of the tin, as the 
 after process is thereby simplified. Supposing, however, 
 that the rough ore has been submitted to fusion with 
 potash and then dissolved, the solution must be treated 
 
 Nur O 
 -> 
 
548 THE ASSAY OF TIN. 
 
 thus : A bar of zinc must be placed in the solution, 
 which will in course of time precipitate tin, copper, and 
 lead ; when all the metals are thus thrown down the zinc 
 is washed and removed, the precipitated metals well 
 washed and dried. To the dried metals strong nitric acid 
 is now to be added, the mass gently heated, and then 
 evaporated to dryness : when cold it is moistened with 
 dilute nitric acid, water added, and the whole filtered. 
 Lead and copper will pass through the filter as soluble 
 nitrates, and the tin will be found in the filter as insoluble 
 peroxide ; this is to be well washed, dried, ignited, and 
 weighed. It contains 78-61 parts of metallic tin. The 
 amount of tin thus obtained, when multiplied by two, will 
 represent the percentage of the ore. 
 
 If, before the action of caustic potash, the ore has 
 been submitted to the action of aqua regia, sulphuretted 
 hydrogen may be passed through the solution of tin chlo- 
 ride, when tin sulphide will be precipitated ; this is to 
 be washed, dried, gently calcined in a platinum crucible 
 until all smell of sulphurous acid has ceased, allowed to 
 cool, reheated with a fragment of ammonium carbonate, 
 as in the case of roasting copper sulphide, and when cold 
 weighed as pure oxide of tin. The calculation for metal 
 is made as above. 
 
 Mr. J. W. B. Hallet has found that tinstone is very 
 easily dissolved by fusion with three or four times its 
 weight of potassium fluoride. The mineral must be finely 
 pulverised. The fused mass is treated directly in the cru- 
 cible with sulphuric acid to expel fluorine, after which, 
 by adding water, filtering, and boiling the filtrate, the whole 
 of the tin is thrown down as stannic acid, which is to be 
 separated from traces of iron in the usual manner. This 
 method of dissolving the ore of tin is much more convenient 
 than fusion with caustic alkalies, or with sulphur and 
 sodium carbonate. 
 
 M. Moissenet precipitates the metal from a solution of 
 the chloride by means of zinc, and then melts the precipi- 
 tated metal in stearic acid. His process comprises five 
 operations : 
 
ASSAY OF TIN IN GUN- AND BELL-METAL. 549 
 
 1. Purification of the ore by treatment with aqua regia. 
 
 2. Eeduction of the residue in the presence of charcoal. 
 
 3. Solution of the tin and iron in hydrochloric acid. 
 
 4. Precipitation of the tin by means of zinc. 
 
 5. Fusion of the precipitate into a button in stearic 
 acid. 
 
 The precipitation of tin by zinc is very rapid, and takes 
 place in strongly acid solutions ; but the amount of acid and 
 the dilution of the chloride influence the condition of the 
 precipitate. In some solutions it appears in brilliant 
 needles, but in very dilute solutions, and always towards 
 the end of an operation, it is only a muddy deposit. The 
 author recommends that a button of zinc be suspended in 
 the liquid by means of a copper wire. When the precipi- 
 tation is finished, the metal is collected and pressed into a 
 porcelain capsule. On applying heat the lump so formed 
 melts in a few minutes if a piece of stearine is added to it. 
 
 Assay of Tin in Gun- and Bell-metal. The follow- 
 ing process was employed for some years in H. Sainte- 
 Claire Deville's laboratory: Dissolve about 5 grms. of the 
 alloy in strong nitric acid contained in a flask provided 
 with a funnel in the neck to prevent loss by spirting. 
 When quite dissolved boil the strong solution for about 
 twenty minutes ; dilute with two or three times its bulk 
 of water, and boil again for the same time. Separate the 
 insoluble tin oxide by decantation or filtration, and weigh 
 after calcining it. (The tin oxide is sometimes rose- 
 coloured, owing to the presence of minute traces of gold ; 
 this may be disregarded.) The nitric acid solution freed 
 from the tin is evaporated on a small platinum or porce- 
 lain dish, and the residue is calcined at a dull red heat. 
 In this manner a mixture of oxides is obtained in a suf- 
 ficient quantity to serve for at least two analyses. 
 
 About 2 grms. of the finely pulverised oxides are 
 placed in a small platinum or porcelain boat, and thence 
 introduced into a small glass tube closed with a good cork 
 suitable for weighing. The boat, the tube, and the cork 
 having been previously weighed, the weight of the oxides 
 is obtained after they have been heated to dull redness in 
 
550 THE ASSAY OF TIN. 
 
 the apparatus, through which a current of dry air circu- 
 lates. After having weighed the whole the current of air 
 is replaced by dry hydrogen, and the tube is heated over 
 a lamp until the contents cease to lose weight. It then 
 contains unreduced zinc oxide together with copper, lead, 
 and iron in the metallic state ; the colour of the product 
 shows the operator when the experiment is concluded. On 
 weighing again the loss of weight indicates with great 
 accuracy the amount of oxygen contained in the oxides of 
 these three metals. 
 
 If the iron and lead are present in inappreciable quan- 
 tities, multiplying this loss by 5 will give very nearly 
 the weight of copper present, and, in consequence, the 
 composition of the alloy itself. 
 
 For bronze, bell-metal, gun-metal, &c., E. Burse 
 ('Zeitschrift fur Analytische Chemie,' 1878, p. 58) pro- 
 ceeds as follows : 1 grm. of the alloy, cut in pieces, is 
 placed in a beaker, covered with 6 c.c. nitric acid of sp. 
 gr. 1-5 : 3 c.c. water are then slowly poured in, and the 
 beaker quickly covered. When the whole is dissolved it 
 is heated to a boil and diluted with 500 c.c. boiling water. 
 The tin oxide, after it is completely settled, is washed with 
 boiling water, and weighed. The filtrate, for the expul- 
 sion of nitric acid, is evaporated with two grms. sulphuric 
 acid, and the copper, after the addition of sulphurous acid, 
 is precipitated with a solution of 2 grm. ammonium sulpho- 
 cyanide. The copper sulphocyanide, after settling, wash- 
 ing, and drying, is weighed as such, or is converted into 
 sulphide. The filtrate is concentrated with the addition 
 of nitric acid, and mixed with ammonia in excess. If 
 iron oxide is deposited, it is filtered off and weighed. The 
 ammoniacal solution of zinc is mixed with ammonium 
 sulphide, avoiding excess ; the sulphide when deposited is 
 filtered off, dried, and, after ignition with sulphur in a 
 current of hydrogen, weighed. 
 
 Estimation of Tin by means of a Standard Solution. 
 The first process to be described is due to M. Gaultier de 
 Clauby, and may be thus performed : The standard solu- 
 tion is made by dissolving 100 grains of iodine in 1 quart 
 
ESTIMATION OF TIN BY A STANDARD SOLUTION. 551 
 
 of proof spirit (specific gravity "920), and is thus standard- 
 ised. Ten grains of pure tin are dissolved in excess of 
 hydrochloric acid, the solution boiled, and allowed to 
 cool : the burette is now filled with the solution of iodine 
 which is gradually added to that of the tin until the former 
 ceases to be decolourised ; as soon, therefore, as the tin 
 solution assumes a faint yellow tinge, sufficient iodine has 
 been added : the quantity thus found sufficient is then 
 noted, and the amount of tin each division of iodine 
 solution is equivalent to is calculated as for iron, copper, 
 and the other standard solutions. 
 
 In the actual assay of tin ore by means of this solution 
 it is necessary that the whole of the tin present be reduced 
 to the state of protochloride : this may be readily effected 
 by boiling the solution of tin for a quarter of an hour with 
 excess of metallic iron, and filtering. To the solution so 
 obtained the iodine is added as above. The tin ore is dis- 
 solved by any of the methods already described. 
 
 M. Lenssen* estimates tin by means of the iodine 
 .solution, but he operates in a liquid containing double 
 potassium and sodium tartrate, and sodium bicarbonate 
 in excess. The results M. Lenssen obtained by this 
 method are satisfactory, by using the atomic weight of tin 
 generally adopted (59). We shall see farther on why 
 M. Lenssen's results agree. 
 
 M. Stromeyer,f having recently occupied himself with 
 the same subject, has succeeded in solving the difficulty. 
 The solution of stannous chloride is carefully introduced 
 into an excess of ferric chloride. The salt of iron becomes 
 reduced to a minimum, according to the following equa- 
 tion : 
 
 2Sn + 2(Fe 2 Cl 6 ) = 2SnCl 2 + 4FeCl 2 . 
 
 It is then estimated by permanganate, as if it were a 
 salt of iron protoxide. The results M. Stromeyer obtains 
 in this way are very exact. The author adds that such a 
 method of estimating is applicable only in the absence of 
 copper or iron, as these two metals decompose potassium 
 
 * Annalen der Chemie und Pharmacie,' vol. cxiv. p. 114. 
 t Ibid. vol. cxvii. p. 261. 
 
552 THE ASSAY OF TIN. 
 
 permanganate as well as the tin ; but it may be of great 
 use in the estimation of commercial salts of tin. 
 
 A method for the analysis of tin ore consists in reduc- 
 ing the finely powdered ore, in a hard porcelain vessel^ 
 in a current of hydrogen gas. A bright red heat must 
 be applied and maintained for a considerable time. The 
 sample is allowed to cool in a current of hydrogen and 
 weighed, the operation being preferably repeated to make 
 sure that the reduction is complete. The loss of weight 
 gives the quantity of stannic acid, tin peroxide (1 part 
 by weight of oxygen representing 4- 673 parts tin per- 
 oxide), if no other oxide is present which may be reduced 
 at the same time. But as tinstone contains almost always 
 iron oxide, part of the loss is due to this compound, so 
 that the crucible, with its contents, should be digested in 
 a beaker with hydrochloric acid. When dissolved, it is 
 diluted, filtered into a large flask, the quantity of the 
 residue (silica, &c.) estimated, the liquid supersatu- 
 rated with ammonia, a sufficient quantity of ammonium 
 sulphide added with flowers of sulphur, so as to convert 
 the stannous sulphide into stannic sulphide, which then 
 dissolves in the ammonium sulphide. The whole is di- 
 gested in the flask, loosely stoppered, till the black iron 
 sulphide is separated from the yellow liquid, filtered under 
 cover, washed with the addition of a little ammonium 
 sulphide, dissolved in hydrochloric acid, oxidised with 
 potassium chlorate, and the ferric oxide (iron peroxide) 
 precipitated with ammonia. The tin may be directly 
 estimated by precipitating the solution of sulphide with 
 dilute sulphuric acid, using a slight excess. The vessel is 
 loosely covered with paper, and very gently heated, till it 
 no longer smells of sulphuretted hydrogen. The yellow 
 tin sulphide is washed upon the filter, dried enough to 
 enable it to be taken out of the funnel along with the 
 paper, and heated very gently, for a considerable time, 
 in a porcelain crucible, at first covered and then open. 
 When the odour of sulphurous acid is no longer percep- 
 tible the contents of the funnel are moistened with a few 
 drops of nitric acid, and are then gradually heated, with 
 
ESTIMATION OF TIN BY A STANDARD SOLUTION. 553 
 
 excess of air, ultimately to full redness. The tin sulphide 
 is thus converted into stannic acid, which is heated for a 
 short time with a little ammonium carbonate to remove 
 every trace of sulphuric acid. 
 
 In another method for the assay of tinstone, the 
 finely ground sample is mixed in a porcelain crucible 
 with 3 parts sodium carbonate, and 3 parts of sulphur, 
 and the mixture is melted, covered, over the, lamp. 
 When cold, the crucible is laid in water, and heat is 
 applied till the mass dissolves, iron or other electro- 
 positive metals remaining as a black sulphide, which is 
 filtered off, washed, and when dry ignited in the air, in 
 order to convert it into oxide. From the dilute alkaline 
 solution the tin sulphide is precipitated by dilute sul- 
 phuric acid, as directed above. 
 
 In case of impure tin ores, the insoluble sulphide may 
 consist of iron, zinc, copper, and bismuth ; whilst the 
 soluble, along with the tin, may contain arsenic, tungsten, 
 and molybdenum. A process for the separation of tin 
 from tungsten has been given at p. 546. 
 
 Tin slags are decomposed by means of aqua regia, and 
 the tin, &c., precipitated by a current of sulphuretted 
 hydrogen. 
 
 Tin furnace products may be in part alloys of tin and 
 iron, lead, tungsten, cobalt, and arsenic. They are finely 
 powdered, digested with aqua regia, the tin thrown down 
 with sulphuretted hydrogen, and the tin sulphide treated 
 as already directed. If the proportion of tungstic acid is 
 considerable, after digestion in aqua regia and diluting, the 
 tungstic acid separates along with some stannic acid. This 
 deposit is filtered off, washed, dried, and repeatedly ignited 
 with sal-ammoniac, in a covered porcelain crucible, till 
 the weight of the residual tungstic acid is constant. The 
 stannic acid is found as difference. Sulphuretted hydro- 
 gen is then passed into the acid solution (from which the 
 tungstic acid has been deposited) ; the remaining tin is 
 thus precipitated as sulphide, and is filtered off. Iron 
 and manganese are estimated in the filtrate. (F. Rarn- 
 melsberg's Quant. Chem. Analysis.) 
 
554 THE ASSAY OF TIN. 
 
 We have seen that M. Stromeyer, by a happy modifi- 
 cation, has reduced the estimation of tin to that of iron. 
 Applying the same principle, a salt of copper may be 
 substituted for a salt of iron. An equivalent quantity 
 of copper can thus be estimated in place of tin ; and 
 M. Mohr's as well as M. Terreil's * experiments show that 
 copper can be very exactly estimated by permanganate of 
 potash. 
 
 A double decomposition takes place on protochloride 
 of tin being added to nitrate or chloride of copper in 
 excess ; a salt of suboxide of copper forms, and the tin 
 passes to the maximum state of oxidation, according to 
 the following equation : 
 
 4CuO + 2SnCl 2 = 2Cu 2 + SnCl 4 + Sri0 2 - 
 
 To estimate tin it is, then, sufficient to transform it 
 into protochloride, to add to it a solution of nitrate of 
 copper slightly in excess, before diluting it with water, 
 and to titrate the liquid obtained by permanganate of 
 potash. 
 
 There are then three different processes for estimating 
 tin by potassium permanganate : 1. To operate with 
 water freed from air by boiling, protecting it from access 
 of air while cooling. 2. To oxidise protoxide of tin in 
 an alkaline medium. 3. To decompose stannous chloride 
 either by a salt of iron, as proposed by M. Stromeyer, or 
 by a salt of copper. 
 
 Alloys of tin and lead, such as solder, inferior tinfoil, 
 and the coating of terne plates, may be treated thus, 
 according to Eammelsberg : 
 
 Oxidise with nitric acid ; evaporate to dryness in the 
 water-bath ; heat the residue rather more strongly ; 
 moisten with nitric acid when cold ; dilute and separate 
 the stannic acid (tin peroxide) by filtration, washing till 
 the filtrate has no longer an acid reaction. When dry 
 the precipitate is detached from the filter, and placed in 
 a porcelain crucible; the paper is burnt on the lid, the 
 
 * Comptes-Rendus, vol. xlvi. p. 230. 
 
ESTIMATION OF TIN BY A STANDARD SOLUTION. 555 
 
 ash added to the contents of the crucible, moistened with 
 a few drops of nitric acid, heated and finally ignited. 
 The tin is calculated from the stannic acid. 
 
 In the filtrate the lead is estimated as sulphate by 
 the addition of sulphuric acid. (See ' Separation of Lead 
 from Copper and Zinc.') 
 
556 
 
 CHAPTEE XIII. 
 
 THE ASSAY OF ANTIMONY. 
 
 ANTIMONIAL substances susceptible of being assayed by the 
 dry way are divisible into two classes. 
 
 CLASS I. In this class are comprised native antimony 
 and all antimonial substances containing oxygen or chlorine, 
 and but little or no sulphur. 
 
 These substances are the following : 
 
 Native antimony, Sb, 
 Antimony oxide, Sb 2 O 3 , 
 Antimonio.us acid, Sb 2 O 4 , 
 Antimonic acid, Sb 2 5 . 
 
 CLASS II. includes antimony sulphide and all antimo- 
 nial ores containing much sulphur. 
 
 Antimony sulphide, Sb 2 S 3 , 
 
 Antimony oxysulphide, Sb 3 + 2Sb 2 S 3 , 
 
 Haidingerite, 2Sb 2 S 3 + 3FeS. 
 
 ASSAY OF ORES OF THE FIRST CLASS. 
 
 Antimony oxides are readily reduced by charcoal, so 
 that their assay presents no difficulty. The assay is con- 
 ducted in precisely the same manner as that of lead oxide ; 
 only, as antimony is much more volatile than lead, the 
 heat must be managed with care, and the assay taken 
 from the fire as soon as finished. When all suitable pre- 
 cautions are taken, the loss of antimony is not very con- 
 siderable ; but Berthier says it is never less than from 5 
 to 6 per cent. This, we think, is too high. The pure 
 protoxide gives 77 per cent, of metal, and antimonious 
 
ASSAY OF ANTIMONY SULPHIDE. 557 
 
 acid 75. The reduction is readily made, without addition, 
 in a charcoal crucible ; but when the substance to be 
 assayed is mixed with impurities, some flux must be 
 added. It succeeds equally well with 3 parts of black 
 flux, with 1 part of tartar, with 1 part of sodium carbo- 
 nate, and 15 per cent, of charcoal, or any other equivalent 
 reducing flux. 
 
 When the substance under assay contains iron oxide, 
 the latter oxide is more or less reduced, and the metallic 
 iron alloys with the antimony. 
 
 Oxidised minerals which contain but a small quantity 
 of sulphur can also be assayed in this manner ; because 
 the sulphide gives up to black flux the small quantity of 
 antimony which it contains, so that but little remains in 
 the slag. The common glass of antimony produces by 
 this method of assay 70 per cent, of antimony, and occa- 
 sionally even more than that. 
 
 The ores of this class occur very seldom, and are only 
 in rare cases subject to assaying. 
 
 ASSAY OF ORES OF THE SECOND CLASS. 
 
 As pure antimony sulphide (antimonium crudum) as 
 well as metallic antimony (regulus of antimony) are mer- 
 cantile substances, the assays of the ores of this class have 
 for their object the estimation of both these bodies. 
 
 1. ASSAY OF PURE ANTIMONY SULPHIDE (ANTIMONIUM 
 CEUDUM). 
 
 Antimony sulphide is almost the only mineral from 
 which antimonium crudum is produced. This mineral 
 generally occurs intermixed with very refractory gangue 
 (gneiss, quartz, limestone, &c.) Antimony sulphide fuses 
 readily at a low red heat, and is not changed during 
 fusion if atmospheric air is precluded. At a white heat it 
 volatilises without change of composition. 
 
 The assay of antimony sulphide is now effected by a 
 liquation process, i.e. by heating the mineral sufficiently in 
 order to melt the antimony sulphide, and, by this means, 
 
558 THE ASSAY OF ANTIMONY. 
 
 to separate it from the refractory gangue. The produc- 
 tion of antimony sulphide on a large scale is done in the 
 same way. 
 
 For the purpose of assaying, two pots or crucibles are 
 used, one standing in the other, and leaving sufficient 
 space between the two to receive the fused antimony 
 sulphide. The bottom of the inside crucible is furnished 
 with holes. The mineral to be assayed is put into the 
 inside crucible, the latter is then closed with a cover, and 
 hermetically luted ; the joints of the two crucibles are 
 also luted. The under crucible is then put on the hearth 
 of a furnace, enclosed with ashes or sand, in order to keep 
 it cool, and the upper crucible, as far as it is outside of 
 the under crucible, is covered with coal, and heated to a 
 moderate red heat. The antimony sulphide will then 
 melt and collect in the under crucible, from which it may 
 be taken out, after cooling, and weighed. 
 
 2. ASSAY OF EEGULUS OF ANTIMONY. 
 
 This assay may be made in two ways : first, by roast- 
 ing and fusing the oxidised r L matter with black flux ; 
 secondly, by fusing the crude ore with iron, or iron 
 scales, with or without the addition of black flux. 
 
 The roasting of antimony sulphide requires much 
 care, for it is very fusible and volatile, as is also the oxide 
 its decomposition gives rise to. The heat ought to be 
 very low during the operation, and the substance con- 
 tinually stirred. When no more sulphurous acid is given 
 off, we may feel [assured that it is perfectly roasted, 
 because no sulphate is ever formed in this operation. 
 
 The roasted sulphide is then fused with three parts of 
 black flux, or its equivalent. 
 
 Metallic iron very readily separates all the sulphur 
 from antimony sulphide ; but as iron sulphide has a 
 specific gravity near that of antimony, the separation is 
 very difficult to manage : a strong fire must be employed 
 when the desulphurisation is complete, to keep the whole 
 body in full fusion, for a considerable time. With these 
 precautions, two buttons are obtained, which separate very 
 
ASSAY OF REGULUS OF ANTIMONY. 55 
 
 well : the one white, and in large plates, which is anti- 
 mony ; and the other a bronze-yellow, a little brighter 
 than the ordinary iron sulphide, because it is mixed with 
 a little metallic antimony. During the operation a con- 
 siderable portion of antimony is always volatilised, which, 
 by this process, is an inconvenience impossible to avoid. 
 
 It is, nevertheless, practised in the large way in some 
 factories ; but a good result is not generally obtained. It,, 
 however, appears that when all the necessary precautions 
 are taken, it can be employed with advantage. 
 
 The first precaution which is indispensable is, mixing 
 with the sulphide only the precise proportion of iron 
 necessary to effect its decomposition, which quantity 
 amounts to about 42 per cent, of its weight. If more be 
 used, the antimony, having a great tendency to play the 
 part of an electro -negative element, will combine with the 
 surplus, and an iron antimonide results, part of which will 
 remain in the antimony and part in the slag. 
 
 Further, the iron ought to be in the finest possible 
 state of division. If the masses be large, a portion of 
 antimony sulphide is volatilised before they can be fully 
 attacked. In general, 63 per cent, of antimony can be 
 extracted from its sulphide by the aid of iron in the small 
 way, but on the large scale it seems that 55 per cent, is 
 the maximum. 
 
 Cast iron cannot be employed instead of wrought,, 
 because sulphur has very little action on it. The desul- 
 phurisation is imperfect, and the slag adheres to the 
 reduced metal. 
 
 One of the greatest inconveniences in separating 
 sulphur from antimony by means of iron is the strong 
 heat necessary to separate the slag from the metal. This 
 may be remedied by making the slag more fusible and 
 less heavy, by the addition of some flux, as an alkaline 
 carbonate or sulphate. 
 
 If antimony sulphide be fused with an alkaline car- 
 bonate and charcoal, a regulus is obtained, and a slag com- 
 posed of an alkaline sulphide and antimony sulphide. If 
 metallic iron be thrown into this slag whilst in fusion, all 
 
560 THE ASSAY OF ANTIMONY. 
 
 the antimony separates immediately, and a new slag is 
 formed as fluid as the former, containing iron sulphide 
 and a sulphide of the alkaline base employed. If, instead 
 of the above process, the iron be mixed intimately with 
 the antimony sulphide and carbonated alkali, the result is 
 the same 100 parts of sulphide, 42 of metallic iron, 50 of 
 sodium carbonate mixed with one-tenth of its weight of 
 charcoal, or 50 of black flux, give 65 to 66 of regulus ; 
 with the same proportion of iron, and only 10 of flux, 
 only 62 per cent, can be obtained. In these two cases the 
 fusion takes place very rapidly and without bubbling, and 
 the slag, which is very liquid, separates readily from 
 the metal. By employing 1 part of alkaline flux, the 
 proportion of iron can be reduced from 2 5 to 30 per cent., 
 and the product of metal is always from 65 to 66 per 
 cent. 
 
 Hence, in making an assay of antimony sulphide, it is 
 always better to employ a smaller quantity of iron than is 
 necessary to complete the desulphurisation, and make 
 up for it by increasing the quantity of flux : then it may 
 be insured that no excess of iron will be present. 
 
 The alkaline sulphates are decomposed into alkaline 
 sulphides by the agency of charcoal at a slightly elevated 
 temperature. The sulphides of the alkaline metals, by 
 combining with the other metallic sulphides, augment 
 their fusibility very considerably. Thus when sodium 
 sulphate, mixed with about one-fifth of its weight of 
 charcoal, is added to a mixture of antimony sulphide 
 and metallic iron, the metallic antimony separates very 
 rapidly, and the slag almost instantly becomes perfectly" 
 fluid. 
 
 But it must be noted that the presence of an alkaline 
 sulphide diminishes the product of regulus, unless the 
 proportion of iron be augmented at the same time. 
 
 For instance, with 
 
 100 parts of antimony sulphide, 
 
 42 parts of iron, 
 100 parts of sodium sulphate, 
 20 parts of charcoal, 
 
ASSAY OF EEGULUS OF ANTIMONY. 561 
 
 but 22 parts of regulus were furnished ; but with 
 
 100 parts of antimony sulphide, 
 42 parts of iron, 
 10 parts of sodium sulphate, 
 2 parts of charcoal, 
 
 62 parts of antimony were easily obtained. 
 
 Instead of metallic iron, pure iron oxide may be used, 
 or any ferruginous matter whatever, provided it is rich ; 
 but it is necessary to add, at the same time, an alkaline 
 flux and charcoal to reduce the iron oxide. 
 
 Not less than 40 parts of iron scales can be employed 
 for 100 of antimony sulphide ; and then, on the addition 
 of 50 to 100 parts of sodium carbonate and 8 to 10 of 
 charcoal, about 56 of regulus are obtained ; but if with 
 100 parts of sodium carbonate from 14 to 15 parts of 
 charcoal be employed, 65 parts of antimony are the result. 
 By augmenting the proportion of scales, that of soda may 
 be diminished. Thus, if from 56 to 60 parts of scales, 10 
 of soda, and 10 of charcoal be employed, 50 parts of 
 regulus are the result ; and if the proportion of soda be 
 50, and that of carbon 10, from 65 to 66, and even 67, 
 parts of regulus are obtainable. 
 
 The fusion always takes place quickly, and the slags 
 are very fluid. 
 
 When antimony sulphide is fused with forge slag 
 (ferrous silicate), sodium carbonate, and charcoal, a very 
 white crystalline regulus, in large plates, is obtained ; to- 
 gether with a bronze-yellow matte, and a black, opaque, 
 vitreous slag, shining like jet, in which the greatest 
 portion of the alkali employed appeared to be concen- 
 trated. 
 
 These three substances separate very readily from each 
 other. 
 
 100 parts of antimony sulphide, 
 80 parts of forge slag, 
 50 parts of sodium carbonate, 
 10 parts of charcoal, 
 
 produced very readily 60 parts of regulus. 
 
 o o 
 
569! THE ASSAY OF ANTIMONY. 
 
 Antimony sulphide may also be analysed by boiling 
 with aqua regia. The residue consists of sulphur and 
 gangue. It is to be washed and dried, then weighed and 
 ignited. The loss will be sulphur, and the remainder is 
 pure gangue. 
 
 Water is then added to the filtered solution, which will 
 cause the precipitation of some of its contained antimony 
 as oxy chloride : this must be separated by filtration. The 
 solution is then to be saturated with potassium carbonate, 
 and a new precipitate will be formed. The solution is to 
 be filtered, and made slightly acid ; then barium nitrate 
 must be added to it to separate its sulphur as barium 
 sulphate, which is to be washed, dried, and weighed ; its 
 weight indicates the amount of sulphur : 116 parts are 
 equal to 16 parts of sulphur. 
 
 The precipitate by water of oxychloride which remains 
 on the filter is redissolved by hydrochloric acid, and its 
 antimony separated in the metallic state by means of zinc. 
 The precipitate formed by potassium carbonate may con- 
 tain lead, copper, iron, and antimony. It must be treated 
 with nitric acid ; this dissolves everything but the antimony, 
 which may then be estimated as antimonic acid. 
 
 It is always best, before conducting the analysis of 
 antimony sulphide, to moisten it with very dilute hydro- 
 chloric acid, in order to dissolve a portion of the calcium 
 carbonate which may form part of the gangue. As 
 the composition of the antimony sulphide is constant, 
 the following process is sufficient in the assay of an 
 antimonial ore : 
 
 Boil the ore, after treatment with dilute hydrochloric 
 acid, with concentrated hydrochloric acid, which dissolves 
 only antimony sulphide, and precipitate the metal as oxy- 
 chloride by means of water. 
 
 Or, after all gangue soluble in dilute hydrochloric acid 
 has been removed, the residue may be weighed, and then 
 acted on by boiling hydrochloric acid, until all action 
 ceases. The residue must be well washed with weak hydro- 
 chloric acid, dried, ignited, and weighed ; the loss of weight 
 corresponds to the percentage of pure antimony sulphide, 
 which contains 72-7 -per cent, of metal. 
 
DETECTION OF ANTIMONY IN SUBLIMATES. 5(33 
 
 Franz Becker (' Zeitschrift flir Analyt. Chemie,' 1878, 
 p, 185) mixes 1 part of the ore with 3 parts sodium car- 
 bonate and 3 parts sulphur, melts in a porcelain crucible, 
 extracts with hot water, decomposes the filtrate with 
 hydrochloric acid, and converts the antimony sulphide 
 into oxide in the usual manner. 
 
 . The following method of estimating antimony is given 
 by Mr. Sutton : 
 
 The oxide of the metal, or any of its compounds, is 
 brought into solution as tartrate by tartaric acid and water ; 
 the excess of acid neutralised by neutral sodium carbonate, 
 then a cold saturated solution of sodium bicarbonate added 
 in the proportion of 20 c.c. to about CM grm. of Sb 2 3 ; to 
 the clear solution starch liquor and -^ iodine are added 
 until the blue colour appears ; the colour disappears 
 after a little time, therefore the first appearance of a per- 
 manent blue is accepted as the true measure of iodine 
 required. 
 
 1 c.c. yV iodine=0-0061 grm. Sb. 
 
 Detection of Antimony in Sublimates. In the examina- 
 tion of mineral bodies for antimony, the test substance is 
 often roasted in an open tube for the production of a 
 white sublimate. Dr. E. Chapman, Professor of Miner- 
 alogy at Toronto, recommends for the detection of anti- 
 mony in this substance the following process a method 
 more especially available when the operator has only a 
 portable blowpipe-case at his command : The portion of 
 the tube to which the chief part of the sublimate is 
 attached is to be cut off by a triangular file, and dropped 
 into a test-tube containing some tartaric acid dissolved 
 in water. This being warmed, or gently boiled, a part 
 at least of the sublimate will be dissolved. Some 
 potassium bisulphate either alone, or mixed with some 
 sodium carbonate and a little borax, the latter to prevent 
 absorption is then to be fused on charcoal in a reducing 
 flame ; and the alkaline sulphide thus produced is to be 
 removed by the point of the knife-blade, and placed in a 
 small porcelain capsule. The hepatic mass is most easily 
 
 o o 2 
 
564 THE ASSAY OF ANTIMONY. 
 
 separated from the charcoal by removing it before it has 
 time to solidify. Some of the tartaric acid solution is 
 then to.be dropped upon it, when the well-known orange- 
 coloured precipitate of antimony sulphide will at once 
 result. 
 
 In performing this test, it is as well to employ a 
 somewhat large fragment .of the test substance, so as to 
 obtain a thick deposit in the tube. It is advisable also to 
 hold the tube in not too inclined a position, in order to 
 let but a moderate current of air pass through it ; and 
 care must be taken not to expose the sublimate to the 
 action of the flame, otherwise it might be converted 
 almost wholly into a compound of antimonious and anti- 
 monic acids, the greater part of which would remain 
 undissolved in the tartaric acid. A sublimate of arsenious 
 acid, treated in this manner, would of course yield a 
 yellow precipitate ; easily distinguishable by its colour y 
 however, from the deep orange antimonial sulphide. The 
 crystalline character, &c., of this sublimate would also 
 effectually prevent any chance of misconception. 
 
 To distinguish Arseniuretted Hydrogen from Antimo- 
 niuretted Hydrogen. On passing a mixture of these two 
 gases through a -tube containing solid pieces of caustic 
 potash, these become covered with a brilliant metallic 
 coating of antimony, whilst the arseniuretted compound 
 escapes undecomposed. A lye of potash, density 1-250, 
 only acts very slightly in a similar case. The fragments 
 of potash which have become metallised by the deposit of 
 antimony are altered in the air ; they soon become white 
 in water, the metallic coating falling to the bottom ; but 
 when it is attempted to collect them on a filter they 
 dissolve before the liquid has even passed through. In 
 the clear filtrate, antimony is found in solution. 
 
 Separation of Tin from Antimony and Arsenic. Dr. 
 Clemens Winckler (' Zeitschrift fur Analyt. Chemie,' 1875, 
 p. 163) proceeds as follows : If the substance is an alloy 
 it is dissolved in a mixture of 4 parts hydrochloric acid, 
 and 5 parts water, with the addition of a sufficient quan- 
 tity of tartaric acid. If the mixed metals exist as sul- 
 
DETECTION OF ANTIMONY IN SUBLIMATES. 565 
 
 phicles they are collected on a filter, washed, and dis- 
 solved upon it in dilute potash lye. The nitrate is mixed 
 with tartaric acid, treated with a current of chlorine, or 
 with bromine in slight excess, and neutralised with hydro- 
 chloric acid. In either case the solution is introduced into 
 a beaker, diluted to 300 or 400 c.c., and so much solution 
 of calcium chloride of a known strength is added that the 
 subsequently precipitated calcium carbonate may exceed 
 the tin present by about 15 times in weight. The liquid 
 is then neutralised with potassium carbonate, potassium 
 cyanide is added, and afterwards a slight excess of potas- 
 sium carbonate, so that the lime may be totally precipi- 
 tated. The liquid is then heated till it begins to boil ; it is 
 allowed to settle, the liquid is poured upon a filter without 
 disturbing the sediment, which is then treated with fresh 
 water, boiled up, allowed to settle, and the clear liquid 
 poured upon the filter. In this manner the bulk of the 
 antimony is removed. 
 
 The precipitate in the beaker is dissolved in a little 
 concentrated hydrochloric acid, tartaric acid is added, 
 the liquid again neutralised with potassium carbonate, 
 and precipitated with potassium cyanide. 
 
 After boiling, the liquid is poured as above through 
 the same filter, three successive portions of water are 
 added, heating each time to a boil, and the precipitate 
 is finally brought upon the filter and completely washed. 
 All the arsenic and antimony are now found in the filtrate, 
 and all the tin, with an excess of calcium carbonate, in the 
 precipitate. 
 
 ASSAY OF ALLOYS OF LEAD AND ANTIMONY (Type Metal). 
 The comminuted alloy is digested in a flask with concen- 
 trated nitric acid till the antimony is oxidised. It is then 
 slightly diluted with water, supersaturated with ammonia, 
 and mixed with an excess of ammonium bi-hydrosulphide, 
 which must be concentrated and yellow. If recently 
 prepared a small quantity of flowers of sulphur is added. 
 The whole is digested in the flask for some time, heating 
 at last almost to a boil. The flask is then corked up and 
 let stand till the deposit, which must be of a pure black, 
 
566 THE ASSAY OF ANTIMONY. 
 
 has subsided, while the supernatant liquid is yellow. If 
 this is not the case more ammonium bi-hydrosulphide is 
 needed. When cold the undissolved lead sulphide is 
 separated by nitration from the solution of the antimony 
 sulphide, washed with cold water, dried, placed in a tared 
 porcelain crucible, the filter burnt separately to ashes, 
 which are added to the precipitate along with some sul- 
 phur, and the whole is ignited in a current of hydrogen. 
 After the crucible with its contents has been weighed the 
 addition of sulphur and the ignition are repeated, in order 
 to find if the weight is constant. The lead is calculated 
 from the weight of the lead sulphide. 
 
 The solution of antimony sulphide is mixed, drop by 
 drop, with hydrochloric acid, stirring all the time till the 
 reaction becomes slightly acid. The beaker is covered 
 with a glass plate and allowed to stand till the antimony 
 sulphide is deposited. It is digested at a very gentle heat 
 till the odour of sulphuretted hydrogen has ceased ; when 
 cold it is filtered upon a filter dried at 120 C., and washed 
 with cold water to which a few drops of hydrochloric 
 acid have been added. When air-dry it is kept at the 
 heat of 120-130C. in the hot-air oven till its weight is 
 constant. A weighed portion of this substance is placed 
 in a bulb- tube and heated in a current of dry carbonic 
 acid, at first very gently, and afterwards at 200-300. 
 The black-grey tin sulphide remains, w r hich is allowed to 
 cool in a current of carbonic acid and weighed. 
 
567 
 
 CHAPTER XIV. 
 
 THE ASSAY OF ZING. 
 
 ALL bodies containing zinc, usually found in the assay 
 office, may be divided into four classes : 
 
 Class I. Zinc ores, in which the metal exists as oxide 
 not combined with silica : 
 
 Earthy zinc oxide, ZnO. 
 
 Manganiferous zinc oxide, ZnO + MnO. 
 
 Zinc aluminate, Gahnite, ZnO,6AL 2 O 3 . 
 
 Franklinite, 3(FeO,ZnO) + (Fe 2 3 ,Mn 2 3 ). 
 
 Anhydrous zinc carbonate, ZnO,CO 2 . 
 
 Hydrated zinc carbonate, ZnO,3H 2 + 3ZnO,C0 2 . 
 
 Class II. Zinc ores, in which the metal exists, as in 
 the former class, as oxide, but partly or wholly combined 
 with silica : 
 
 Anhydrous zinc silicate. 
 
 Hydrated zinc silicate, electric calamine. 
 
 Class III. Zinc ores, in which the metal is partly or 
 wholly combined with sulphur. 
 
 Zinc sulphide (blende, Black Jack), ZnS. 
 Zinc oxysulphide. (This is rare.) 
 Zinc sulphate, ZnO, S0 3 ,7H 2 0. 
 Zinc selenide, ZnSe. (Very rare.) 
 
 Class IV. Alloys. 
 
 ASSAY OF ORES OF THE FIEST CLASS. 
 
 In order to reduce the zinc oxide contained in sub 
 stances of this class, it is sufficient to mix them with char- 
 coal and expose them to a white heat. 
 
568 THE ASSAY OF ZINC. 
 
 At the moment of reduction the zinc is in a vaporised 
 state. Its vapours, however, are readily condensable, so 
 that the operation may be conducted in an ordinary retort, 
 and all the metal is deposited in the neck without loss. 
 It seems from this that nothing is so easy, at first sight, 
 as the assay of zinc oxide ; but it is not so. It is very 
 easy to reduce all the oxide, but it is not so easy to col- 
 lect all the zinc ; nor is it easy to condense it all in the 
 metallic state, and in consequence to estimate the pre- 
 cise proportion in the ore submitted to assay. 
 
 This difficulty consists, first, in the deposit being 
 extended over a large surface and often adhering very 
 strongly to the sides of the retort, so that it is nearly im- 
 possible to detach it ; and, secondly, as the neck of the 
 retort is open, the air, having access to it, brings to the 
 state of oxide all the vapour nearest the end of the neck. 
 The proportion of zinc oxidised is larger in proportion to 
 the smallness of the quantity submitted to assay, and is 
 always very considerable where no more than 200 to 400 
 grains are operated upon. 
 
 It is not, therefore, in the extraction of the zinc from 
 its oxide that the assay is rendered partially uncertain, but 
 in its collection. 
 
 The distillation of zinc requires a very high tempera- 
 ture, and cannot be performed in retorts of glass ; those of 
 earthenware must be employed. It is not necessary to 
 lute these retorts when they are of good quality ; and 
 they are better thin, because they heat more rapidly, and 
 are not so likely to crack. 
 
 After the mixture of oxide and charcoal has been 
 introduced into the retort it is placed in the fire. The 
 neck ought to have a long tube of glass, with a narrow 
 bore adapted to it, so as to collect all the zinc which may 
 escape from the wide part of the neck of the retort. 
 This disposition is also convenient, as it does not allow 
 such a free access of air. 
 
 It is heated gradually until it is white inside ; the zinc 
 is reduced and volatilised, and condensed in the neck : the 
 greater the heat, the nearer the orifice. The metal can be 
 
ASSAY OF OEES OP THE FIRST CLASS. 569 
 
 detached readily from the neck, if it be well blackleaded 
 inside. It is necessary, from time to time, to observe the 
 state of the neck, because when very narrow it is often 
 obstructed, and, if not cleaned out with an iron rod, an 
 explosion might be caused. 
 
 When the operation is finished the apparatus is allowed 
 to cool, the retort taken out, placed carefully on its side 
 and broken, in order that any particles of zinc which have 
 condensed in its dome may be removed. 
 
 If the approximate proportion of metallic zinc alone be 
 required, all is collected and fused at a very gentle heat in 
 a crucible with some black flux ; but if the true quantity 
 of zinc is to be estimated it must be done in a more exact 
 manner. The deposit must be collected with all possible 
 care ; the neck must then be broken to pieces, and every 
 piece having adhering to it either zinc or oxide must be 
 placed on one side, and digested in hot nitric acid, which 
 takes up those substances. If any be in the glass tube, it 
 must be carefully cleansed by means of acid, and the solu- 
 tion added to that produced by the digestion of the broken 
 neck, and the deposit mechanically collected, in nitric acid. 
 The solution is then evaporated gradually to dryness and 
 heated to redness. The nitrate, by these means, is decom- 
 posed and transformed into oxide, four-fifths of the weight 
 of which is equal to the quantity of metallic zinc produced 
 in the assay. 
 
 The foregoing is the method of estimation by dis- 
 tillation ; the following is the method of estimation 
 by difference. Two plans of assay in this manner may 
 be adopted : first, at an ordinary assay temperature ; 
 secondly, at a very high temperature, as that of an iron 
 assay. In all cases it is necessary to commence with the 
 expulsion of all volatile bodies the ore may contain. If 
 water or carbonic acid alone be present, simple calcination 
 will do ; but if carbonaceous matter, roasting must be had 
 recourse to. 
 
 When the assay is made at an ordinary assay tempera- 
 ture, the sample is finely pulverised and mixed with from 
 15 to 20 per cent, of equally finely pulverised charcoal, 
 
570 THE ASSAY OF ZINC. 
 
 and pressed into a crucible, on which a cover is placed, 
 but not luted, and rapidly heated to whiteness. When no 
 more zinc vapour is disengaged it is cooled, and the mixture 
 in the pot collected. The residue ought to be pulverulent ; 
 but as it is mixed with some charcoal, it is roasted, and 
 then weighed. It is evident that the loss represents the 
 zinc oxide. The charcoal added, it is true, leaves a small 
 quantity of ash, but it is too small to be worth accounting 
 for. 
 
 In making the assay in the manner described, it is to be 
 feared that a small quantity of the oxide might remain un- 
 decomposed, and that a part of the residue might adhere to 
 the crucible, and could not be detached ; and, lastly, there 
 is always a degree of uncertainty as to the state of oxida- 
 tion in which the iron will exist after roasting. No incon- 
 venience of this nature presents itself when the assay is 
 made at a very high temperature. This mode is the most 
 exact of all, and leaves nothing to be desired. 
 
 The assays of zinc at a high temperature are made 
 exactly as those of iron. They are made in a charcoal 
 crucible, with the addition of fixed fluxes, suitable to effect 
 the fusion of the gangues mixed with the zinc oxide, if 
 they be not fusible by themselves. The button is weighed ; 
 it is a compound of slag and grains of iron, which are col- 
 lected and their weight ascertained, and, by the difference, 
 that of the slag. The weight of oxygen which the iron 
 has lost during its reduction is then added to it, and by 
 deducting from the substance the weight of the button 
 and the oxygen so obtained we have the proportion of 
 zinc oxide reduced in the assay. On the other hand, by 
 deducting from the weight of the slag the weight of flux 
 added, the weight of earthy substances and irreducible 
 oxides which were mixed with the zinc oxide is ascer- 
 tained. 
 
 These results can be shown in a tabular form, in the 
 following manner : 
 
 Let m be the weight of the crude ore, n the weight of 
 the calcined ore, r the weight of the flux added, f the 
 Aveight of the cast iron, s the weight of the slag, o the 
 
ASSAY OF OKES OF THE FIRST CLASS. 571 
 
 weight of oxygen combined with the iron, calculated from 
 the weight of metal produced, ' z the weight of the zinc 
 oxide, then: 
 
 in crude ore = calcined ore . n 
 
 r fixed fluxes ...... r 
 
 n + r 
 
 Gives metal . . /\ Tfttfll f , >> 
 Gives slag . . i/S^ }/+*** 
 
 Zinc oxide n + r f s o 
 Flux added . . . r 
 
 Earthy matter . s - r 
 
 The following is an actual experiment by Berthier : 
 
 100 crude ore = calcined ore ..... 83'3 
 
 10 kaolin (china clay) acted on by acids . . . lO'O 
 
 7 marble = lime ....... 4-0 
 
 97-3 
 
 Gave metal . 45-3 \ Tfl fi1 .o^ 
 
 Gave slag. . 16-3 j ^ W*j*<M . . 80-7 
 
 Zinc oxide . . 16'6 
 Fluxes added ...... . . 14-0 
 
 Earthy matters ....... 2-0 
 
 The above result was confirmed by wet analysis, show- 
 ing at once the exactitude of the process. 
 
 Estimation of Amount of Zinc by the Wet Process in 
 Ores of the First Class. Dissolve 50 grains of the finely 
 pulverised ore in nitric acid, evaporate to dryness, allow 
 to cool. Digest the cold mass with a little dilute nitric 
 acid, gently warming during the digestion ; add water, and 
 then filter. To the filtered solution add excess of caustic 
 ammonia, gently warm, and filter. The excess of caustic 
 ammonia dissolves the zinc oxide which it at first threw 
 down, as well as any manganese oxide that may be present. 
 This solution containing the zinc, and probably manganese, 
 must be separated from the precipitate produced by the 
 ammonia by filtration, the insoluble matter in the filter 
 washed with water containing a little ammonia, and the 
 washings so obtained added to the first strong filtrate. If 
 no manganese be present, ammonium sulphide may now be 
 added to the filtered liquid until it produces no further 
 white precipitate of zinc oxide. The liquid and precipitate 
 must now be allowed to stand in a warm place for about 
 
572 THE ASSAY OF ZINC. 
 
 an hour, then filtered, and the zinc sulphide on the filter 
 washed with water containing a little ammonium sulphide. 
 After a few washings it is to be dissolved in dilute hydro- 
 chloric acid, and, if necessary, the solution filtered. To 
 the filtered solution is added excess of sodium carbonate ; 
 zinc carbonate is thrown down, which in its turn is col- 
 lected on a filter, washed, dried, separated from the filter, 
 ignited, and weighed. Four-fifths of its weight is metallic 
 zinc. If, by previous experiments by blowpipe or other- 
 wise, manganese were found to be present, the ammoniacal 
 solution containing the mixed oxides must be thus treated. 
 Excess of acetic acid is to be added to it, and a stream 
 of sulphuretted hydrogen gas passed through it until no 
 further precipitation takes place ; by this means the whole 
 of the zinc is deposited as sulphide, whilst the manganese 
 remains untouched in the liquid. The zinc sulphide is to 
 be collected on a filter and treated with hydrochloric acid, 
 &c., as just described. 
 
 ASSAY OF OKES OF THE SECOND CLASS. 
 
 Zinc silicates are not reducible by charcoal alone ; but 
 when in contact with substances which have the property 
 of combining with silica, they are reduced completely, 
 even at a moderate temperature. All the modes of assay 
 just described for ores of the first class apply to those of 
 the second, with the exception that the flux, instead of 
 being merely reducing, must have a true fluxing property 
 also : lime or magnesia -are good fluxes. 
 
 Wet Estimation of Zinc in Ores of the Second Class. 
 Ores of this class are best decomposed by strong hydro- 
 chloric acid with a small admixture of nitric acid. When 
 thoroughly decomposed, and the solution evaporated to 
 dryness, the residue is moistened with hydrochloric acid, 
 and treated exactly as described for Ores of the First Class. 
 
ASSAY OF ORES OF THE THIRD CLASS. 573 
 
 ASSAY OF ORES OF THE THIRD CLASS. 
 
 In order to assay the substances containing sulphur 
 which belong to this class, they must be roasted, and then 
 treated as the ores of the first and second class. Zinc 
 sulphide may be roasted without difficulty ; and when 
 the operation is made with care, the roasted ore contains 
 neither sulphur nor sulphuric acid. The only precaution 
 necessary to observe is, that the heat must be carefully 
 regulated at first, in order to avoid fusion which might 
 take place, especially when a certain amount of iron sul- 
 phide is present. Towards the end the heat may be in- 
 creased to decompose any sulphate that may be formed. 
 Both a reducing and fusing substance must be added in 
 this case as in the last, in order to effect the fusion of 
 the gangue. 
 
 Wet Estimation of Zinc in Ores of the Third Class. 
 These ores are to be finely pulverised, treated with strong 
 nitric acid, at first with a gentle heat ; and, lastly, boiled 
 until the sulphur separates in bright yellow transparent 
 globules, as described under the Wet Assay of Copper 
 Ores of the Second Class. The solution so obtained 
 is to be evaporated to dryness, moistened with hydro- 
 chloric acid, and treated as described for ores of the first 
 class. 
 
 If ores of this class, or of either of the two former, con- 
 tain copper, they must be thus treated : 
 
 The ore is to be decomposed by an appropriate acid, 
 evaporated to dryness, moistened with hydrochloric acid r 
 water added, and the solution filtered. A current of sul- 
 phuretted hydrogen gas is now to be passed through the 
 solution until, even after violent agitation, it smells strongly 
 of it. It is now to be filtered, and the black precipitate 
 on the filter contains all the copper as copper sulphide, 
 that substance being insoluble in dilute acid, whilst in a 
 solution acidulated with either of the strong mineral acids 
 as nitric, hydrochloric, or sulphuric zinc is not at all 
 acted on by sulphuretted hydrogen. The solution, now 
 
574 THE ASSAY OF ZINC. 
 
 freed from copper, is placed in an evaporating basin and 
 boiled for about a quarter of an hour ; nitric acid is then 
 added to per oxidise all the iron present, and the solution 
 allowed to cool. When cold, the zinc is separated by 
 means of ammonia, and the ammoniacal solution treated 
 as already described. 
 
 Assay of Cupriferous Blendes. 
 
 Mr. E. Monger has described the following rapid and 
 accurate process for the assay of cupriferous blendes : 
 
 Fifteen grains of the blende are taken and treated with 
 aqua regia and evaporated ; the residue is re-moistened 
 with hydrochloric acid, and re-evaporated, dissolved in 
 water with a drop or two of hydrochloric acid, 10 to 
 
 20 c.c. of ammonia added, and the iron filtered off and 
 washed. 
 
 The filtrate, which is now about 200 to 250 c.c. in 
 volume, is acidified with hydrochloric acid, heated, and 
 placed in a porcelain dish ready for buretting with ferro- 
 cyanide of O'Ol strength. To the solution of zinc and 
 copper add sodium sulphide solution enough to precipitate 
 the whole of the copper and leaving a little excess ; then 
 proceed with the buretting, with ferrocyanide and uranium 
 acetate as indicator. 
 
 In some experiments Mr. Monger took a non-cupri- 
 ferous blende and weighed out six samples, into three of 
 which he put some sulphate of copper solution, equal to 
 
 21 per cent, copper in the blende, and proceeded with 
 them as above explained, and obtained perfectly concordant 
 results. 
 
 FOURTH CLASS. ALLOYS. 
 
 The alloys of zinc with iron, copper, and tin may 
 be assayed by heating them to whiteness for about an 
 hour in a charcoal crucible with an earthy flux (calcium 
 silicate is the best), and weighing the resulting button : 
 the loss will be nearly equivalent to the quantity of zinc 
 present. 
 
VOLUMETRIC DETERMINATION OF ZINC. 575 
 
 The Wet Estimation of Zinc in Substances of the Fourth 
 Class. These substances are treated precisely as described 
 under the heads Wet estimation of Zinc in First, Second, 
 and Third Classes. 
 
 VOLUMETRIC ESTIMATION OF ZINC. 
 
 Galetti 's Process. -A good method of volumetrically 
 estimating the amount of zinc in ores is given in the 
 ' Zeitschrift fiir analytische Chemie ' for 1869, by Maurizio 
 Galetti. Chief Assayer at the Eoyal Assay Office, Genoa. 
 The following is a description of the process : Suppos- 
 ing zinc sulphide (blende) is to be assayed, about half a 
 gramme of the finely pulverised ore is to be treated with 
 concentrated nitric acid, and boiled to incipient dry ness, 
 until the sulphur left undissolved does not contain any 
 particles of undissolved ore. Then add strong hydrochloric 
 acid, and boil again until no nitric acid is left. Calamine 
 (zinc carbonate) should at once be acted upon with hydro- 
 chloric acid ; but, in order to make sure of the complete 
 oxidation of all the iron the ore may happen to contain, 
 it is best to add to the acid a few decigrammes of pure 
 potassium chlorate. After having boiled this solution for 
 a few minutes it is diluted with distilled water ; a large 
 excess of ammonia is added to the solution, which is then 
 boiled and slightly acidified with acetic acid. After brisk 
 agitation boil again for a few minutes, and then super- 
 saturate with ammonia. The liquid is then poured out of 
 the flask into a suitable glass vessel, and the flask is rinsed 
 out with a sufficient quantity of distilled water to bring 
 the bulk of the fluid up to half a litre. This having been 
 done, the fluid is very cautiously and gradually acidified 
 with dilute acetic acid, one part acid sp. gr. 1-07 to 10 of 
 distilled water. Any large excess of this should be avoided, 
 as the solution should only be very slightly acid. As soon 
 as the basic iron acetate has subsided, the precipitation of 
 the zinc by means of a standard solution of potassium 
 ferrocyanide may be proceeded with. 
 
 The ferrocyanide solution is made by dissolving 41-25 
 
576 THE ASSAY OF ZINC. 
 
 grms. of the salt in as much distilled water as will make 
 the solution weigh exactly one kilogramme. 
 
 The presence of compounds of lead (as, for instance, 
 lead carbonate, sulphate, or sulphide) occurring along with 
 the ores of zinc does not interfere with the completeness 
 of the precipitation of zinc as zinc ferrocyanide. This even 
 holds good up to 10 per cent, of metallic lead. Since some 
 ores of zinc, especially calamine, often contain manganese, 
 it is best to add to the ammoniacal solution, before any 
 acetic acid is added, a few drops (from 2 to 4) of bromine, 
 in order to convert the manganese protoxide into proto- 
 sesquioxide, leaving the solution standing for twenty-four 
 hours after the addition of the bromine. 
 
 The ammoniacal solution of zinc chloride being colour- 
 less, there should be added to it, previous to cautious 
 acidification by means of dilute acetic acid, a few drops 
 of tincture of litmus, in order to more readily hit the 
 precise point of sufficient acidification, which is known by 
 the blue coloration changing to a rose-red. 
 
 The zinc ferrocyanide which is mixed with iron oxide 
 preserves its naturally white colour as long as the liquid 
 contains free zinc, but its colour changes to a greyish 
 white as soon as a very slight excess of the ferrocyanide 
 standard solution is present ; the liquid also then becomes 
 turbid, and the precipitate settles very slowly. By these 
 characteristic signs the end of the operation may be always 
 recognised. In order to make sure, the liquid should be 
 touched with a glass rod which has been just previously 
 moistened with a dilute solution of ammoniacal copper 
 nitrate ; this will have the effect of indicating any excess of 
 the ferrocyanide solution, by producing the more or less 
 intense colour characteristic of copper ferrocyanide. The 
 zinc solution should be at a temperature of from 40 to 50, 
 whereby the rapid subsidence of the zinc ferrocyanide is 
 promoted. 
 
 Filtration is not necessary, as the presence of the gela- 
 tinous silica (due to the decomposition of zinc silicates 
 occurring in the ores of that metal) does not interfere 
 with the correctness of this method of estimating zinc 
 quantitatively. 
 
VOLUMETRIC ESTIMATION OF ZINC. 577 
 
 Fresenius * gives the following methods for the volu- 
 metric estimation of zinc : 
 
 1. SCHAFFNER'S METHOD^ MODIFIED BY C. KUNZEL^; AS EMPLOYED 
 IN THE BELGIAN ZINC- WORKS ; DESCRIBED BY C. GROLL. 
 
 a. Solution of the Ore and Preparation of the Ammoniacal 
 
 Solution. 
 
 Powder and dry the ore. 
 
 Take O5 grm. in the case of rich ores, 1 grm. in the 
 case of poor ores, transfer to a small flask, dissolve in 
 hydrochloric acid with addition of some nitric acid, by the 
 aid of heat, expel the excess of acid by evaporation, add 
 some water, and then excess of ammonia. Filter into a 
 beaker, and wash the residue with lukewarm water and 
 ammonia till ammonium sulphide ceases to produce a 
 white turbidity in the washings. The zinc oxide remain- 
 ing in the hydrated ferric oxide is disregarded. Its quan- 
 tity, according to Groll, does not exceed 0-3 0-5 per cent. 
 This statement probably has reference only to ores contain- 
 ing relatively little iron ; where much iron is present the 
 quantity of zinc left behind in the precipitate may be not 
 inconsiderable. The error thus arising may be greatly 
 diminished by dissolving the slightly washed iron precipi- 
 tate in hydrochloric acid, and adding excess of ammonia. 
 But the surer mode of proceeding is to add to the original 
 solution after evaporating off the greater part of the 
 free acid as above, and allowing to cool dilute sodium 
 carbonate nearly to neutralisation, then to precipitate the 
 ferric oxide with boiling sodium acetate, filter, and 
 wash. The washings, after being concentrated by evapo- 
 ration, are added to the nitrate, and the whole is then 
 mixed with ammonia till the first formed precipitate is 
 re-dissolved. 
 
 If the ore contains manganese provided approximate 
 results will suffice digest the solution of the ore in acids, 
 
 * 4th English edition, p. 653, published by Churchill and Sons. 
 t ' Journ. f. prakt. Chem.' 73, 410. t Ibid. 88, 486. 
 
 ' Zeitschrift f. anal. Chem.' 1, 21. 
 
 P P 
 
578 THE ASSAY OF ZINC. 
 
 after the addition of excess of ammonia and water, at a 
 gentle heat, for a long time, and then filter off, with the 
 iron precipitate, the hydrated manganese protosesquioxide 
 which has separated from the action of the air. The safer 
 course though undoubtedly less simple is, after sepa- 
 rating the iron with sodium acetate, to precipitate the 
 manganese by passing chlorine through, or by adding 
 bromine and heating. 
 
 If lead is present, it is separated by evaporating the 
 aqua regia solution with sulphuric acid, taking up the 
 residue with water and filtering ; then proceed as directed.* 
 
 bi Preparation and Standardising of the Sodium Sulphide 
 
 Solution. 
 
 The solution of sodium sulphide is prepared either by 
 dissolving crystallised sodium sulphide in water (about 
 100 grm. to 1,000-1,200 water), or by supersaturating a 
 solution of soda, free from carbonic acid, with sulphuretted 
 hydrogen, and subsequently heating the solution in a flask 
 to expel the excess of sulphuretted hydrogen. Whichever 
 way it is prepared, the solution is afterwards diluted, so 
 that 1 c.c. may precipitate about O'Ol grm. zinc. Prepare 
 a solution of zinc by dissolving 10 grm. chemically pure 
 zinc in hydrochloric acid, or 44-122 grm. dry crystallised 
 potassium and zinc sulphate in water, or 68-133 grm. dry 
 crystallised potassium and zinc sulphate in water, and 
 making the solution in either case up to 1 litre with water. 
 
 Each c.c. of this solution corresponds to 0-01 grm. 
 zinc. Now measure off 30-50 c.c. of this zinc solution 
 into a beaker, add ammonia till the precipitate is re-dis- 
 solved, and then 400-500 c.c. distilled water. Eun in 
 sodium sulphide as long as a distinct precipitate continues 
 to be formed, then stir briskly, remove a drop of the fluid 
 on the end of a rod to a porcelain plate, spread it out so 
 
 * Concerning the direct treatment of roasted zinc ores with a mixture of 
 carbonated and caustic ammonia, comp. E. Schmidt (Journ. f. prakt. Chew. 
 51, 257). By this treatment the zinc oxide, which was combined with car- 
 bonic acid, is dissolved, whilst that combined with silicic acid is, for the most 
 part, left undissolved. 
 
VOLUMETRIC ESTIMATION OF ZINC. 579 
 
 that it may cover a somewhat large surface, and place in 
 the middle a drop of pure dilute solution of nickel chloride. 
 If the edge of the drop of nickel solution remains blue or 
 green proceed with the addition of sodium sulphide, testing 
 from time to time, till at last a blackish grey coloration 
 appears surrounding the nickel solution. The reaction is 
 now completed, the whole of the zinc is precipitated, and 
 a slight excess of sodium sulphide has been added. The 
 precise depth of colour of the nickel must be observed and 
 remembered, as it will have to serve as the stopping signal 
 in future experiments. To make sure that the zinc is 
 really quite precipitated, you may add a few tenths of a 
 c.c. more of the reagent, and test again ; of course the 
 colour of the nickel drop must be darker. Note the 
 number of c.c. used, and repeat the experiment, running 
 in at once the necessary quantity of the reagent less 1 c.c., 
 and then adding 0'2 c.c. at a time till the end-reaction 
 is reached. The last experiment is considered the more 
 correct one. The sodium sulphide solution must be re- 
 standardised before each new series of analyses that is, 
 if it is kept in bottles containing air ; if, on the contrary, 
 oxygen is excluded by passing the air through an alkaline 
 solution of pyrogallic acid previously to its entering the 
 bottle, the solution would without doubt keep unaltered. 
 
 c. Estimation of Zinc in the Solution of the Ore. 
 
 Proceed in the same way with the ammoniacal solution 
 prepared in a as with the known zinc solution in b. Here 
 also repeat the experiment, the second time running in at 
 once the required number of c.c., less 1 of sodium sulphide, 
 and then adding 0*2 c.c. at a time till the end-reaction 
 makes its appearance. The second result is considered 
 the true one. There are three different ways in which 
 this repetition of the experiment may be made. You may 
 either weigh out at the first two portions of the zinc ore, 
 or you may -weigh out double the quantity required for 
 one experiment, and make the ammoniacal solution up to 
 1 litre, and employ -J litre for each experiment ; or, lastly, 
 
 r P 2 
 
580 THE ASSAY OF ZINC. 
 
 having reached the end-reaction in the first experiment, 
 you may add 1 c.c. of the known zinc solution, which will 
 destroy the excess of sodium sulphide, and run in sodium 
 sulphide in portions of O2 c.c. till the end-reaction is 
 again attained. Of course, in this last process to obtain 
 the second result, you deduct from the whole quantity of 
 sodium sulphide used the amount of the same, correspond- 
 ing to 1 c.c. of the zinc solution. 
 
 If the ore contain copper, which frequently occurs in 
 the case of blendes, determine by a preliminary experi- 
 ment the number of c.c. of sodium sulphide which are 
 necessary to precipitate the copper, and, at the completion 
 of the zinc analysis, deduct them. In this case, let the drop 
 to be tested with nickel solution pass through a small 
 filter on its way to the porcelain plate, in order to avoid 
 the injurious influence of the copper sulphide on the nickel 
 reaction. If, however, the copper amounts to more than 
 2 per cent., remove it from the acid solution by sulphu- 
 retted hydrogen, evaporate the filtrate with nitric acid, 
 dilute, treat with ammonia, and estimate the zinc as 
 above. 
 
 In careful hands the error will, according to C. Kttnzel, 
 never exceed ^ per cent. 
 
 d. Further Modification of the Process. 
 
 To ascertain the point when the whole of the zinc is 
 precipitated and the sodium sulphide begins to predomi- 
 nate, Schafmer* employed flocks of hydrated ferric oxide, 
 which he produced by the addition of a few drops of ferric 
 chloride to the ammoniacal zinc solution, and which settled 
 at the bottom ; while Barreswil f used small pieces of white 
 porcelain, which were covered with ferric chloride, and 
 thrown into the ammoniacal zinc solution. Sodium sul- 
 phide is added till the flocks or the pieces of porcelain turn 
 black. In neither case is the end-reaction so exact as with 
 nickel solution. 
 
 * ' Journ. f. prakt. Chem.' 73, 410. 
 
 t ' Journ. de Pharm.' 1857, 431 ; ' Polyt. Centralbl.' 1858, 285. 
 
VOLUMETRIC ESTIMATION OP ZINC. 581 
 
 With the help of lead-paper, however, the point may 
 be hit with great precision. Moisten a piece of white 
 filter-paper with solution of lead acetate, place it on a 
 layer of blotting-paper, drop some ammonium carbonate 
 upon it, so as to form a thin coating of lead carbonate, let 
 the blotting-paper absorb the excess of moisture, and then 
 spread the lead-paper on a porcelain plate. As soon as 
 you imagine the zinc to be nearly all precipitated, lay a 
 small piece of filter-paper on the lead-paper, and then dip 
 the end of a blunt glass rod in the fluid, and press it some- 
 what gently on the small piece of filter-paper. When the 
 sodium sulphide begins to be in excess, a brown spot forms 
 on the lead-paper. Fr.'Mohr* applies the lead reaction in 
 another manner. He makes an alkaline solution of lead 
 by warming together lead acetate, Eochelle salt, and solu- 
 tion of soda ; he first places a drop of this on filter-paper, 
 and then close by a drop of the precipitated zinc solution, 
 so that the circle formed by the spreading of the solution 
 to be tested may cut the circle of the lead solution. As 
 soon as the sodium sulphide begins to predominate, the 
 portion of the circumference of the lead circle which lies 
 in the other circle turns black. 
 
 2. H. SCHWARZ'S METHOD, f 
 
 Prepare an amnioniacal solution as in 1, a. 
 
 Heat gently, and mix with a moderate excess of am- 
 monium sulphide. Allow the precipitated zinc sulphide 
 to subside, then filter, using a tolerably large ribbed filter 
 of rapidly filtering paper, moistened with boiling water, 
 and warming the fluid to accelerate the operation, which 
 would otherwise require considerable time. Wash the 
 precipitate with warm water mixed with a little ammonia, 
 until the last drops no longer blacken a solution of lead 
 oxide in soda. 
 
 Transfer the filter with the precipitate to a beaker, add 
 
 * Mohr's * Lehrbuch der Titrirmethode,' 2 Aufl. 377. 
 
 t See his * Anleitung zu Maassanalysen,' Nachtrage, p. 29 (Brunswick). 
 Compare also v. Gellhorn ('Chem. Centralbl.' 1853, 291), who has made many 
 .analyses by Schwarz's method. 
 
582 THE ASSAY OF ZINC. 
 
 a dilute solution of slightly acidified ferric chloride, cover 
 with a close-fitting glass plate, and let the mixture stand 
 for ten minutes ; then heat gently. Under these circum- 
 stances the zinc sulphide decomposes completely with the 
 ferric chloride to zinc chloride, ferrous chloride, and sul- 
 phur : Fe 2 Cl 6 + ZnS =ZnCl 2 + S + 2FeCl 2 . 
 
 Now add sulphuric acid, and heat gently until the 
 sulphur has agglutinated. Filter, and wash the filter, and 
 estimate the iron in the fluid as protochloride by per- 
 manganate,* 2 eq. iron correspond to 1 eq. zinc. If the 
 quantity of zinc sulphide is not very great, the filter may 
 be broken, and the zinc sulphide washed into a flask 
 which already contains the solution of ferric chloride. 
 The great objection to this method lies in the washing of 
 the zinc sulphide, which, as is well known, is a long and 
 troublesome operation. A possible loss of sulphuretted 
 hydrogen on mixing the zinc sulphide with ferric chloride 
 may be prevented by conducting the decomposition in a 
 flask, connected with a U-tube containing ferric chloride. 
 
 3. CARL MOHR'S METHOD, f 
 
 This method is based upon the following considera- 
 tions : 
 
 I. If a solution of zinc acetate, acidified with acetic 
 acid, is mixed with an excess of pota'ssium ferrocyanide, 
 the whole of the zinc is thrown down in the form of a 
 reddish-yellow precipitate of zinc ferrocyanide, Zn 3 (Cy 6 Fe 2 ). 
 
 II. If solution of potassium iodide is now added in ex- 
 cess, we have this decomposition : 2[Zn 3 (Cy 6 Fe 2 )] + 2KI + 
 2( A,HO) = 3[Zn 3 (Cy 3 Fe)] + 2(KO, A) + H 2 (Cy 3 Fe) + 21. 
 
 III. 1 eq. liberated iodine corresponds, accordingly, to- 
 3 eq. zinc. 
 
 IV. If potassium iodide is made to act upon zinc ferro- 
 cyanide in a neutral fluid, the liberated iodine acts upon 
 the potassium ferrocyanide present in that case, which 
 leads to the formation of a little potassium ferrocyanide ; 
 
 * Without doubt the ferric chloride might be replaced by ferric sul- 
 phate, by which means the presence of hydrochloric acid would be avoided. 
 t Dingler's ' Polyt. Journ.' 148, 115. 
 
CARL MOHRS METHOD. 583 
 
 the remaining free iodine, therefore, will not indicate, with 
 accuracy, the quantity of zinc present. But whereas the 
 reaction actually takes place in acid solution of zinc ace- 
 tate, as above directed, it may be assumed that potassium 
 acetate and free hydroferrocyanic acid are formed ; and as 
 iodine exercises no appreciable action upon the latter sub- 
 stance, the iodine liberated in the process indicates, with 
 tolerable accuracy, the amount of zinc present. 
 
 The process is as follows : 
 
 Treat the ore with aqua regia, as in 1, a, and drive off 
 the greater part of the free acid ; nearly neutralise with 
 sodium carbonate, add sodium acetate in excess, boil, 
 filter, and wash with boiling water mixed with a little 
 sodium acetate. The solution is free from iron : it con- 
 tains the whole of the zinc, but, in presence of manganese, 
 also the whole of the latter metal. Hence the process is 
 not applicable in the presence of manganese. 
 
 Mix the solution of zinc, prepared as directed, with 
 potassium ferrocyanide in slight excess, i.e. until a 
 sample of the clear supernatant fluid gives a blue precipi- 
 tate with a ferrous salt. Then add a sufficient quantity 
 of potassium iodide. The fluid acquires a brown colour, 
 in consequence of the liberation of iodine ; the white pre- 
 cipitate of zinc ferrocyanide is suspended in the brown 
 fluid, 
 
 Now estimate the free iodine by means of sodium 
 hyposulphite, and calculate 3 eq. zinc for each eq. iodine. 
 The results obtained by C. Mohr are very satisfactory. 
 The method can be employed only if the acetic acid solu- 
 tion contains no other heavy metal besides zinc, and, more 
 particularly, no manganese. 
 
 For estimating the value of zinc powder, J. Drewson 
 proposes the following method : 
 
 He prepares two solutions, the one of pure fused 
 potassium dichromate say 40 grms. per 1,000 c.c. and 
 the other of crystalline ferrous sulphate, about 200 grms. 
 in 1,000 c.c. The iron solution must be strongly acidu- 
 lated with sulphuric acid to prevent oxidation. In order 
 to find the respective value of the two liquids, 10 c.c. of 
 
584 THE ASSAY OF ZINC. 
 
 the iron solution are measured into a beaker, a little sul- . 
 phuric acid is added, and the other solution is dropped 
 in from a burette until a drop of the mixture is no longer 
 turned blue by potassium ferrocyanide. About 1 grm. of 
 the zinc powder is then weighed, placed in a beaker with 
 100 c.c. of the chromic solution ; 10 c.c. of dilute sul- 
 phuric acid are added, the whole is well stirred, 10 c.c. 
 more of the sulphuric acid are added, and allowed to act 
 for about a quarter of an hour, with diligent stirring. 
 When everything is dissolved except a small insoluble 
 residue, an excess of sulphuric acid is added, and 50 c.c. 
 of the iron solution, in order to reduce the greater part of 
 the excess of chromate ; more of the iron solution is then 
 added from a burette, till a drop displays a distinct blue 
 reaction with ferrocyanide, and the mixture is then titrated 
 back with chromate till this reaction disappears. From 
 the total number of c.c. of the iron solution consumed, 
 the quantity is deducted which corresponds to the ferrous 
 solution employed. The chromate contained in the re- 
 mainder, if multiplied by 0'66113, shows the metallic zinc 
 contained in the powder. 
 
 In order to separate copper from zinc by a single 
 precipitation with sulphuretted hydrogen, G. Larsen 
 (' Zeitschrift fur analyt. Chemie,' 1878, p. 312) passes 
 sulphuretted hydrogen into the solution, filters, washes 
 the precipitate first with hydrochloric acid of sp. gr. 
 1*05, through which sulphuretted hydrogen has been 
 passed, and then with pure sulphuretted hydrogen water. 
 Both the precipitation and the washing are effected by 
 heat. 
 
 Brass, pinchbeck, false gold-leaf, bronze not containing 
 tin, &c., are dissolved in nitric acid as directed for alloys 
 of silver and copper. 
 
 The acid liquid is diluted with water and a current of 
 sulphuretted hydrogen is introduced, the vessel being kept 
 covered with a glass plate. This is continued till the pre- 
 cipitate has subsided and the liquid becomes clear and 
 colourless. The precipitate copper sulphide is poured 
 upon a filter, and the filtrate collected in an evaporating 
 
SEPARATION OF COPPER FROM ZINC. 585 
 
 basin. To prevent oxidation the filtration must be per- 
 formed rapidly, and the air excluded as far as possible by 
 keeping the funnel and the beaker covered with a glass 
 plate. The filter should be kept constantly full, till it 
 has received all the precipitate. 
 
 All particles adhering to the sides of the beaker and 
 to the gas delivery tube are washed into the filter by 
 means of a feather and the washing- bottle. The precipi- 
 tate is then washed on the filter moderately with cold 
 water, a few drops of sulphuretted hydrogen water being 
 added each time. The precipitate is dried, transferred to 
 a tared porcelain crucible, upon the lid of which the filter 
 is reduced to ashes, which are added to the precipitate. 
 The whole is mixed with a few centigrms. of flowers of 
 sulphur, and the crucible is covered with a peculiar lid, in 
 which a bent porcelain tube opens, through which dried 
 hydrogen is introduced into the crucible, which is then 
 placed over a lamp, and as soon as the common air is 
 expelled from the apparatus it is heated to redness. The 
 contents are then let cool in the current of hydrogen. 
 Thus is obtained pure cuprous sulphide, Cu 2 S, which is 
 weighed, and from its weight that of the copper is calcu- 
 lated. 
 
 The filtrate containing the zinc is concentrated by 
 evaporation to expel excess of sulphuretted hydrogen. 
 Crystallised sodium carbonate is then gradually added, 
 with stirring till the effervescence ceases, the precipitation 
 is effected, and the liquid has an alkaline reaction. In 
 order to lose nothing by spirting, the beaker is kept 
 covered with a glass plate, which is afterwards rinsed into 
 the beaker by means of the washing-bottle. The liquid is 
 brought to a boil, filtered, and the precipitate well washed 
 with boiling water. It is dried and strongly ignited, when 
 it becomes converted into zinc oxide. The filter is burnt 
 on the lid of the crucible. The zinc oxide is weighed, and 
 from it the zinc is calculated. 
 
 If lead is also present, the solution of the metals is 
 placed in an evaporating dish, and mixed with so much 
 moderately concentrated sulphuric acid as to convert the 
 
586 THE ASSAY OF ZINC. 
 
 lead into sulphate and leave an excess, which is afterwards 
 chiefly expelled by evaporation. When cold, water and 
 a little alcohol are added ; the precipitate is poured upon 
 a filter, dried at 120, washed with cold alcoholic water, 
 dried, and weighed. 
 
 The filtrate is treated with sulphuretted hydrogen, as 
 for the estimation of copper, &c. (Rammelsberg). 
 
587 
 
 CHAPTEE XV. 
 
 THE ASSAY OF MERCURY. 
 
 MERCURY is found in the native or metallic state, and as 
 sulphide or cinnabar : 
 
 Native mercury, Hg. 
 
 Mercury sulphide, cinnabar, Hg 2 S. 
 
 Bituminous mercury sulphide. 
 
 There are other minerals of mercury met with, but 
 hitherto not in sufficient quantity to be worked for the 
 metal. They are 
 
 Zinciferous mercury subsulphide. 
 
 Zinciferous mercury sulphide. 
 
 Mercury selenide. 
 
 Mercury subchloride. 
 
 Mercury iodide. 
 
 Silver amalgam (see Silver). 
 
 Assay of Mercurial Ores. The estimation of mercury 
 is generally made by distillation. When the mercury 
 is present in the form of native mercury, or mercury 
 oxide, it is distilled without any addition. The ore (say 
 from 500 to 1,000 grains) is placed in an iron or earthen- 
 ware retort, which is set over a suitable fire, and the 
 heat raised gradually, and kept up, until the whole of 
 the mercury has passed over. The mercury which passes 
 over is collected either in the neck of the retort or in a 
 receiver fitted for that purpose such as a glass flask kept 
 cool by affusion with water. When but a small quantity 
 is operated upon (say 150 to 200 grains), it is most con- 
 venient to use a glass retort, or bent tube retort, heating 
 it gradually over a charcoal fire, taking care to keep the 
 
588 THE ASSAY OF MERCURY. 
 
 upper part so hot that no metallic mercury may adhere to 
 it. It must be heated nearly to the melting-point of the 
 glass, and until all the mercury has come over. 
 
 When the operation is finished, the neck is cut off, 
 weighed, the mercury detached, and weighed again : the 
 loss of weight is the amount of mercury. Or the metal 
 may be detached by means of a feather, and allowed to 
 fall into a basin of water, which, if heated for a few 
 seconds, will cause the mercury to collect into one glo- 
 bule : the water may be decanted, and the mercury dried 
 at the ordinary temperature, and weighed. 
 
 The mercury wholly condenses in the neck of the 
 retort, under the form of a metallic dew. Some may by 
 chance pass off; but in order to prevent such an occur- 
 rence, the beak of the retort is plunged into water, or a 
 loose plug of linen, moistened with water, is introduced 
 into the neck, the end of which is plunged into water, by 
 which means the neck of the retort is kept constantly cool, 
 and the mercury is found deposited on the linen, from 
 which it may be detached by shaking in water. 
 
 When large quantities of substances containing mer- 
 cury are operated upon, it is necessary to heat very 
 strongly towards the end, in order that the centre of the 
 mass may receive a sufficient amount of heat to effect its 
 decomposition. Naked glass retorts cannot be used, and 
 either coated glass or porcelain retorts must be employed. 
 In the large way, as in the distillation of amalgams, &c. 5 
 cast-iron retorts are used. 
 
 As before stated, all substances containing mercury, 
 either in its metallic state or as oxide, are distilled without 
 addition, but with the others it is necessary to employ 
 some reagent which will separate and retain the sulphur, 
 selenium, &c. ; this reagent may be a metal, as iron, 
 copper, or tin ; or black flux, or a mixture of quicklime 
 and charcoal : iron filings are most often used. For 
 cinnabar about 50 per cent, of iron filings is required, in 
 order to prevent any of it being sublimed ; the actual quan- 
 tity required is only about 24 per cent., but an excess is 
 necessary, in order, as before stated, to prevent loss : 50 
 
ASSAY IN THE DEY WAY. 58$ 
 
 per cent, of iron filings may be employed for the selenides, 
 &c. When black flux is used, from about 50 to 70 per 
 cent, is employed. Caustic lime may be employed in the 
 proportion of 30 per cent, mixed with 30 per cent, of its 
 weight of charcoal. After the ore to be assayed is care- 
 fully mixed with any of the above fluxes, it is always 
 advisable to cover it, w r hen in the retort, with a thin 
 layer of the flux employed, in order to avoid all chance 
 of any loss. 
 
 In estimating mercury by distillation it is necessary,, 
 especially if the metal is in the state of chloride or sul- 
 phide, to take certain precautions, without which a 
 portion of the sulphide or chloride would volatilise 
 without decomposition. H. Eose gives the following 
 directions for carrying out the operation : Introduce 
 into a glass tube capable of resisting fusion, closed at one 
 end, and measuring from 35 to 50 centimetres in length, 
 a column of sodium bicarbonate, then one of quicklime, 
 and then a w r ell-blended mixture of the mercurial com- 
 pound and quicklime, and, finally, a column of quick- 
 lime. The open end of the tube is drawn out and bent 
 round so as to enter a small flask containing water. The 
 tube is heated as if for an organic analysis, commencing 
 at the open end and finishing with the sodium bicar- 
 bonate. The operation ended, cut the bent end of the 
 tube. Collect all the mercury in the flask, dry with paper, 
 and afterwards over sulphuric acid, and then weigh it. 
 
 The quicklime should not be replaced by hydrate of 
 lime. That would occasion all the inconvenience of an 
 analysis of a sulphuretted combination of mercury. The 
 water acting on the calcium sulphide would form sul- 
 phuretted hydrogen, which, by dissolving in the water of 
 the receiver, would, in time, transform a portion of the re- 
 duced mercury into sulphide. It is advisable in some cases 
 to replace the sodium bicarbonate by magnesium carbonate- 
 
 Combinations containing mercury iodide are riot en- 
 tirely decomposed when treated as above. Biniodide and 
 protoiodide are condensed in the extremity of the tube 
 simultaneously with the metallic mercury. 
 
590 THE ASSAY OF MERCURY. 
 
 To analyse these combinations recourse must be had 
 to metallic copper, the operation being similar to that 
 with quicklime. 
 
 Berthier, who experimented with an ore containing 
 arsenic, realgar, &c., and cinnabar from Huanca-Velica, 
 in Peru, found, after very many fruitless experiments, the 
 following method best adapted to its examination for 
 mercury : 
 
 The ore was heated in a retort with four to five times 
 its weight of litharge. From the litharge, the arsenic 
 sulphide, &c., a fusible slaggy mass was formed, while 
 the cinnabar was decomposed into sulphurous acid and 
 metallic mercury. The mercury volatilised completely at a 
 moderate heat, and collected in the fore part of the neck 
 of the retort and in the receiver. The single precaution 
 which must be observed for the success of the assay con- 
 sists in only gradually and moderately heating the clay or 
 glass retort, in order to prevent its being perforated by 
 the corroding effect of the litharge before the operation 
 is ended. 
 
 If the assay sample is extremely poor in mercury, the 
 ordinary assay method becomes somewhat inconvenient 
 and uncertain, on account of the large quantity which 
 must then be subjected to distillation in the assay. For 
 this case Berthier found it more appropriate to digest the 
 assay sample with aqua regia, wash it thoroughly, evapo- 
 rate the whole mass of fluid to dryness, and then treat the 
 dry mass, which contains all the mercury as chloride, 
 further in the dry way. He found that if mercury 
 chloride (corrosive sublimate) is heated with litharge, it 
 volatilises without undergoing ' any change. If, besides 
 the litharge, coal-dust is also added, or if instead of it 
 metallic lead is used in great excess, the chloride is 
 reduced to subchloride, which volatilises, but not the 
 smallest drop of mercury is thus produced. The best 
 reducing agent for the mercury chloride contained in the 
 dry mass is black flux, of which three parts by weight 
 are used. Since the mass to be subjected to distillation has 
 been greatly diminished by the treatment of aqua regia, 
 
ASSAY IN THE DRY WAY. 591 
 
 and the subsequent evaporation, and no high heat is now 
 required for the decomposition, the distillation may be 
 performed in a glass retort. When the gangue in the 
 poor ore is calcium carbonate, all the lime must be dis- 
 solved out by moderately strong acetic acid before the 
 treatment with aqua regia. 
 
 By this method the smallest trace of mercury in an ore 
 or amalgamation product can be shown and estimated 
 by its weight. 
 
 EschkcCs process for assaying mercury ores is given 
 as follows in the 4 Chemical News ' for July 1872. The 
 method may be used for cinnabar, mercuriferous fahlerz, 
 &c. The ore should be weighed in a balance turning 
 with one milligramme. The quantity of ore for the assay 
 varies according to its richness, as follows : 
 
 Ore containing up to 1 per cent. .. .10 grammes 
 
 1 10 ,, . . . o 
 
 10 ,,30 . . .2 
 
 ,, over 30 . . .1 gramme 
 
 The ore is introduced into a porcelain crucible, the edge 
 of which has been ground flat, and mixed with about 
 half its weight of clean iron filings by means of a glass 
 rod, and is then evenly covered with a layer of iron filings 
 about J to f inch thick. A concave cover, made of fine 
 gold, about two inches in diameter and 12 to 15 grammes 
 in weight, is now placed on the crucible after having been 
 carefully weighed ; the concavity is filled with distilled 
 water and the crucible placed on a triangle and heated for 
 ten minutes by a Bunsen burner or Argand spirit-lamp, 
 during which time the mercury is volatilised and deposits 
 itself on the gold. The gold cover is then removed, the 
 water poured off, and the mirror of mercury on the convex 
 side washed with alcohol from a wash -bottle. After being 
 dried in the water-bath, the cover is allowed to cool tho- 
 roughly, and is then weighed in a balance turning with -J 
 milligramme with 50 grammes in the pan. The increase 
 of weight gives the quantity of mercury in the ore. 
 During the weighing the cover is placed on an empty 
 
592 THE ASSAY OF MERCURY. 
 
 porcelain crucible. The mercury is then driven off by 
 heating the cover gently in the flame of a Bunsen burner 
 or a spirit-lamp, in a place where there is a good draught,, 
 and the empty crucible and cover are subjected to a 
 second weighing as a check. In order that the assay 
 should succeed the following conditions must be fulfilled : 
 The cover must fit closely, so as to avoid loss of 
 mercury, and must be deep enough to hold a sufficient 
 quantity of water to keep it cool. The iron filings must 
 be free from grease, which would prevent the proper 
 formation of the mirror ; the washing with alcohol must 
 not be omitted, as it removes all the bituminous sub- 
 stances which spoil the mirror, and assists the drying ; 
 it must be dried in a water-bath for two or three minutes, 
 cooled in the desiccator, and weighed when fully cool. 
 When assaying rich ores the alcohol used in washing 
 the cover must be collected, as it may contain a little 
 amalgam ; it must be poured into the concavity of the 
 cover, which will take up any little globules of mercury. 
 The most exact results are obtained in the case of poor 
 ores containing less than 10 per cent. 
 
 Assay for the Amount of Cinnabar in an Ore. The ore 
 to be assayed is distilled, without addition, in a glass retort, 
 and the sublimed cinnabar collected and weighed. The 
 ores containing mercury combined with sulphur are often 
 mixed with bituminous matters and calcium carbonate : 
 then, when an assay is to be made for cinnabar, it often 
 happens that a portion of it is decomposed, either by the 
 carbon present or by the aid of the bituminous matter and 
 lime, and a little metallic mercury is driven off with the 
 cinnabar. In this case, having weighed the mixture of 
 cinnabar and mercury, the mixture is treated with nitric 
 acid, which dissolves only the latter, and pure cinnabar 
 remains, the weight of which is taken, and the quantity of 
 mercury dissolved ascertained by the difference, and from 
 that the quantity of cinnabar calculated which that quan- 
 tity of mercury would yield. Every 86 parts of mercury 
 furnish about 100 of cinnabar. 
 
 If the gangue of the ore be fixed in the fire, the assay 
 
ASSAY IN THE WET WAY. 593 
 
 may be made by mere calcination, and the loss of weight 
 will correspond either to the metallic mercury, oxide, or 
 sulphide it may contain. 
 
 For the estimation of mercury in the wet way in 
 its usual ores, which are mixtures of cinnabar with lime- 
 stone, clay, iron oxide, and bituminous matters, the sample 
 is first treated with hydrochloric acid, which dissolves the 
 lime, &c< The liquid is then poured off and the residue 
 digested with aqua regia, when the mercury is dissolved 
 as chloride. The solution is filtered off from the insoluble 
 residue, the greater part of the free acid removed by 
 evaporation, diluted and heated with a solution of sul- 
 phurous acid, to reduce the ferric oxide to the ferrous con- 
 dition. Without previous filtration it is then treated with 
 sulphuretted hydrogen, the precipitate of mercury sulphide 
 is brought upon a weighed filter, and dried at 100-110 C. 
 Any free sulphur present may be removed by digestion 
 with a solution of sodium hyposulphite. The mercury 
 sulphide may be afterwards tested for the presence of 
 other volatile metals. 
 
 If it is intended to weigh mercury as mercurous 
 chloride (calomel) after reduction with stannous chloride, 
 the ore should be dissolved not in aqua regia, but in a 
 mixture of hydrochloric acid and potassium chlorate. 
 Free chlorine is expelled by heat, and the liquid is placed 
 in a flask and mixed with a clear solution of stannous 
 chloride, to which hydrochloric acid has been added, and 
 the whole is boiled for a few moments. When cold, the 
 liquid is decanted off, the mercury rinsed in a crucible, 
 washed with acid water, and dried in the desiccator. 
 
 In amalgams mercury is generally estimated by 
 placing the sample in a porcelain crucible and heating in 
 a current of hydrogen gas, when the mercury is volatilised. 
 (Rammelsberg . ) 
 
 (This method, however, is not applicable to the amal- 
 gams of the alkaline metals.) 
 
 QQ 
 
594 THE ASSAY OF MERCURY. 
 
 Electrolytic Estimation of Mercury. 
 
 Don Luis de la Escosura has described two processes 
 for the estimation of mercury by electrolysis. In the first 
 process he operates upon about 15 grains of ore, treated 
 with 20 c.c. of water mixed with 10 to 15 c.c. of hydro- 
 chloric acid. The whole is gently heated in a porcelain 
 capsule, and when the liquid is on the point of boiling 
 from 15 to 30 grains of potassium chlorate in powder are 
 added by small portions. When the action is over 50 c.c. 
 of water are added to the liquid, and it is boiled afresh 
 until the odour of chlorine is no longer perceptible. At 
 this moment 20 to 30 c.c. of a saturated solution of 
 ammonium sulphite are added, boiling again for a few 
 minutes ; the capsule is then withdrawn from the fire, and 
 let settle. Care must be taken to supply the water which 
 is lost by evaporation. The addition of the ammonium 
 sulphite is to precipitate selenium and tellurium before 
 electrolysing the liquid. After settling for half an hour 
 the liquid is filtered, and the insoluble residue is washed. 
 The volume of the filtrate should be about 200 c.c. This 
 liquid is put in a glass beaker, and the electrodes are 
 introduced. These are plates of metal ; the one may be 
 of platinum, but the other which communicates with the 
 zinc pole of the battery must be of pure gold. The plates 
 are suspended vertically in the liquid. The battery con- 
 sists of two Bun sen elements. In twenty-four to thirty 
 hours the operation is .completed ; the mercury is depo- 
 sited upon the gold plate. The increase of weight in this 
 corresponds to the quantity of mercury contained in the 
 weight of metal taken. 
 
 Second Process. Direct Electrolysis of the Mercury Ore 
 without Previous Solution. The ore, in fine powder, is 
 placed in a platinum capsule, in a mixture of water, 
 hydrochloric acid, and ammonium sulphite. For a 10 per 
 cent, ore the following proportions are employed : Ore, 
 6 grains ; acid, 10 c.c. ; water, 20 c.c. ; sulphite, 20 c.c. ; 
 so as to make up about 120 c.c. These proportions are 
 
ELECTROLYTIC ESTIMATION OP MERCURY. 595 
 
 modified according to the richness of the ore. The 
 platinum capsule, 3^ inches in diameter, is placed on 
 a support, and a disc of gold is plunged into it ; 
 soldered to a rod of gold, communicating with the 
 zinc pole of the battery, the capsule is connected to 
 the other pole. In twenty-four hours the operation is 
 complete : the mercury is deposited upon the disc of gold, 
 the increase of weight indicating the quantity of the 
 metal. 
 
 In preference to a Bunsen battery, Don Luis de la 
 Escosura uses six elements formed of a cylindrical vessel 
 of glass, 0*1 metre in diameter and 0*16 metre in height, 
 in which a zinc disc is suspended by means of two copper 
 wires at two-thirds of the height of the glass above the 
 bottom. At the centre of the zinc disc is a copper wire 
 bent twice at right angles, so that the other extremity of 
 the wire plunges into the glass vessel of the next element. 
 To increase the surface of contact it is well to make the 
 copper wire end in a spiral. The copper wire immersed 
 in the liquid, except the spiral, is coated with an insu- 
 lating substance. 
 
 The glass is filled with plain water after the crystals of 
 copper sulphate have been introduced, and in a few hours 
 the battery is ready for use. 
 
 This second process, though much simpler than the 
 first, is strictly accurate, even in the case of very poor 
 ores containing less than 0*1 per cent, of mercury. The 
 accompanying impurities do not interfere. 
 
 VOLUMETRIC ESTIMATION OF MERCURY. 
 
 The process we have found most trustworthy is that of 
 M. J. Personne, described in the ' Comptes Eendus,' Ivi. 63, 
 as follows. The author says : ^ 
 
 ' The process at which I have arrived, after many 
 fruitless attempts, is founded on a well-known fact that 
 a combination of mercury iodide with potassium iodide, 
 forming the double iodide, gives a colourless solution. 
 Thus, two solutions in equal quantities, one containing one 
 
 Q Q 2 
 
,596 THE ASSAY OF MERCURY. 
 
 equivalent of mercury bichloride, the other two equiva- 
 lents of potassium iodide, being mixed, by pouring the 
 mercurial solution into that of the potassium iodide, mer- 
 cury iodide will be produced by the contact of the two 
 solutions, which dissolves in proportion to its formation, 
 until the mercurial solution added is equal in volume to 
 that of the alkaline iodide used. The slightest excess of 
 bichloride causes the formation of a persistent red pre- 
 cipitate, giving the liquid a very perceptible red tint 
 even by artificial light. This coloration, which indicates 
 that the saturation is complete, gives to this mode of 
 estimation a precision and nicety quite as great as that 
 of litmus, used to ascertain the saturation of an acid by 
 a base. The mercurial solution must always be poured 
 into the alkaline iodide not the alkaline iodide into 
 the mercurial solution; otherwise, though the last reac- 
 tion may be the same, it is impossible to obtain exact 
 results, because the mercury iodide produced, not being 
 brought simultaneously with its formation (in a nascent 
 state) into contact with the alkaline iodide with which it 
 is to combine, becomes sufficiently cohesive to dissolve 
 but slowly in the potassium iodide. Thus, in operating 
 with the same liquids, the quantity of alkaline iodide 
 which must be added to dissolve the mercury iodide pre- 
 cipitated varies according to the time employed in effect- 
 ing the estimation, and that in considerable proportions. 
 I have no doubt that it is through operating in this way 
 that potassium iodide has hitherto been rejected as a 
 medium for the exact estimation of mercury. 
 
 ' Two normal liquids are necessary to effect this esti- 
 mation. 
 
 6 1. Normal Standard Solution of Potassium Iodide. 
 Obtained by dissolving 3 3 '20 grammes of pure potassium 
 iodide in water enough to make 1 litre of solution. 10 
 cubic centimetres of this solution represent O'l gramme of 
 metallic mercury. 
 
 4 2. Normal Standard Solution of Mercury Bichloride. 
 Prepared by dissolving 13*55 grammes of mercury bi- 
 chloride in water, so as to make 1 litre of solution. The 
 
VOLUMETRIC ESTIMATION OF MERCURY. 597 
 
 solution of mercurial salt is facilitated by the addition of 
 5 equivalents, or 30 grammes, of sodium chloride, which 
 has no influence on the reaction, like all neutral alkaline 
 salts ; 10 cubic centimetres of this solution also represent 
 O'l gramme of mercury. Of these 10 centimetres, divided 
 into 100 parts, each division represents O'OOl gramme of 
 mercury. This mercurial solution serves to test the purity 
 of the alkaline iodide solution or to give the standard of 
 an unknown solution. 
 
 ' Liquids ten times more diluted may be prepared 
 without injuring the nicety of the reaction or the exact- 
 ness of the results ; fractions of a milligramme may thus 
 be estimated. 
 
 ' The estimation is effected in the following manner : 
 10 cubic centimetres of a normal solution of iodide being 
 
 o 
 
 measured into a small saturating vessel, pour into it, 
 constantly shaking the vessel, the solution of bichloride, 
 measured in Guy-Lussac's burette. If the two liquids are 
 pure, it will require exactly 100 divisions of the burette 
 before the light red tint appears in the saturated liquid to 
 indicate the close of the operation. When the mercurial 
 solution is weak a proportionally larger quantity must be 
 added. If, on the other hand, it is too strong, less must 
 be added. As will be perceived, this is very similar to 
 the chlorometric process. 
 
 ' This new method of estimating mercury being appli- 
 cable only to bichloride, it became desirable to extend its 
 application to a greater number of mercurial compounds, 
 if not to all. This side of the question presented diffi- 
 culties not easily resolved in a satisfactory manner. It 
 was, in fact, necessary to transform all the mercurial 
 compounds into a perfectly neutral solution of bichloride. 
 I was obliged to set aside successively the use of aqua 
 regia, and even of hypochlorous acid. The great volatility 
 of mercury bichloride, even in a boiling solution, caused 
 too great a loss. M. Eivot's process that is to say, the 
 action of chlorine in a solution of hydrate of potash or 
 soda is perfectly successful. Take, for instance, the 
 estimation of mercury in cinnabar. Eeduce one gramme 
 
598 THE ASSAY OF MERCURY. 
 
 of cinnabar to a fine powder. Weigh it on paper, and 
 introduce it into a matrass. Pour into the matrass 20 
 cubic centimetres of a caustic soda solution, with which 
 mix the paper and its contents by quickly shaking ; then 
 send a current of chlorine, which need not be washed, 
 into the liquid. The action of the chlorine produces a 
 slight heat, which is gradually brought to boiling-point, 
 by which time all the matter will have disappeared. To 
 insure success, the temperature must be carefully managed 
 at the commencement. If it is raised too quickly, part of 
 the matter remains undissolved. The solution being com- 
 plete and saturated with chlorine, it is kept boiling long 
 enough to expel all the excess of chlorine. The boiling 
 may be prolonged without incurring any loss of bichloride, 
 which is not volatile in presence of alkaline chloride. 
 The solution when cooled is poured into a graduated 
 tube. The matrass as well as the tube for conducting the 
 chlorine is washed two or three times with water, and 
 the washing added to the original liquid, so as to form 100 
 cubic centimetres of solution. I effected the estimation 
 with the standard solution of iodide, of which 10 cubic centi- 
 metres represent 1 gramme of mercury. To saturate these 
 10 cubic centimetres it required 115 divisions of the chloro- 
 mercurial solution. These 115 divisions contain then 0*1 
 gramme of mercury. Now, as all the mercury contained in 
 the analysed cinnabar is spread through the 10,000 divi- 
 sions of solution, we have the quantity of mercury found 
 by the experiment by means of a simple proportion.' 
 
 Mr. G. Attwood gives the following process for the 
 quantitative blowpipe assay of mercury. The compounds 
 to be assayed may be divided into three classes. Class A, 
 containing metallic mercury, cinnabar, tiemanite, sub- 
 oxide, protoxide, and mixed sulphides. Class B, calomel, 
 corrosive sublimate, and iodide of mercury. Class C 5 
 amalgams of gold, silver, copper, lead, zinc, tin, &c. 
 
 Class A. 10 to 20 grains of the ore, finely powdered 
 and passed through a sieve, 2,000 holes to the square inch, 
 are mixed with 5 to 10 times their weight of powdered 
 litharge and distilled over a spirit-lamp in a small glass 
 
BLOWPIPE ASSAY OF MERCURY. 599 
 
 retort, 1^ inch long and J inch in diameter. To this 
 retort is fitted, by means of a cork, a glass tube, slightly 
 curved, 2^ inches long and fths of an inch in diameter. 
 The end of this tube dips under water contained in a 
 small porcelain crucible. The operation lasts only a few 
 minutes. The mercury is carefully collected from the 
 glass tube and crucible. The retort is broken' up and its 
 contents carefully powdered and examined by a lens for 
 mercury. The globules are then united by gently warm- 
 ing under water, and the dry mercury weighed. 
 
 Class B. A quantity of the finely powdered ore, equal 
 to 10 grammes, is mixed with three times its volume of 
 potassium oxalate and one volume of potassium cyanide. 
 The apparatus closely resembles that used in Class A, but 
 the retort has a small bulb. 
 
 Class C. These amalgams are sometimes powdered 
 with difficulty, and it is often advantageous to add a 
 known weight of pure mercury, so as to render them 
 semi-fluid before distilling. 10 to 30 grammes of the amal- 
 gam are usually taken for an assay. A turned steel retort 
 is used for distillation, which is effected in a small charcoal 
 furnace heated by a blowpipe flame ; the head of the 
 retort is accurately ground to fit over the body. The 
 retort, including the cup and cap, is 1 inch high ; the 
 neck of the cap is 2 inches long. The paper contains 
 full-size illustrations of the different retorts, &c., which 
 are made by Casella. (See 'Journal of the Chemical 
 Society/ 1879.) 
 
GOO 
 
 CHAPTEE XVI. 
 
 THE ASSAY OF SILVER. 
 
 ALL argentiferous substances may be divided into two 
 classes, as follows : 
 
 CLASS I. Minerals containing silver : 
 
 Silver glance (AgS), containing 87 per cent, of Ag. 
 
 Brittle silver ore (6AgS.SbS 3 ), containing 70-4 per cent, of Ag. 
 
 Light red silver ore (3AgS,AsS 3 ), containing 65-4 per cent, of Ag. 
 
 Dark red silver ore (3AgS,SbS 3 ), containing 59 per cent of Ag. 
 
 Light and dark fahlerz (argentiferous grey copper ore), containing from 
 
 5-7 to 18-31-8 per cent, of Ag. 
 
 Argentiferous copper sulphide (Cu 2 S,AgS), containing 53 per cent, of Ag. 
 Polybasite 9(Cu 2 S,AgS) + (SbS 3 ,AsS 3 ), containing 72-94 per cent of Ag. 
 Slags. 
 
 Cupel bottoms. 
 Dross. 
 Litharge, &c. 
 
 CLASS II. Metallic silver and alloys, either native or 
 otherwise. 
 
 General Observations on the Assay of Ores and Substances 
 of Class No. 1. 
 
 In order to separate silver from this class of sub- 
 stances, an alloy of the precious metal with lead must 
 be formed. The different methods by which this object 
 can be attained are the following : first, fusion with a 
 reducing flux ; secondly, fusion with oxidising reagents ; 
 thirdly, scorification. 
 
 All substances containing lead in the state of oxide, 
 such as carbonates, phosphates, &c., are fused directly 
 with a reducing flux, as also are slags, old cupels, litharge, 
 &c. All plumbiferous sulphides, &c., containing silver, 
 
THE ASSAY- OF SILVER. 601 
 
 are assayed as for lead by the processes already pointed 
 out, taking care to follow the method which gives the 
 largest proportion of lead. 
 
 All argentiferous minerals containing copper may be 
 assayed as copper ores ; because an alloy of copper and 
 silver can be cupelled by means of lead. 
 
 In making assays of silver with lead or copper, it is 
 sometimes necessary to commence the operation by roast- 
 ing the ore ; under other circumstances, also, argentiferous 
 matters are roasted. 
 
 There is nothing very particular to be observed in 
 this roasting ; the temperature alone requires attention by 
 managing well at the commencement of the operation, in 
 order to avoid softening, and especially to avoid a very 
 rapid disengagement of arsenical vapours, because a very 
 considerable amount of silver may be lost by that means. 
 
 All substances which contain reducible oxides are 
 fused with a reducing flux, as also those from which char- 
 coal separates metals which alloy with lead, or metals 
 which do not hinder the process of cupellation ; but it is 
 necessary to add to the reducing flux a certain proportion 
 of litharge, in order to produce metallic lead, with which 
 the silver may alloy. A mixture of metallic lead and any 
 suitable flux may be substituted for that of litharge and a 
 reducing flux ; but the latter is preferable, because the 
 lead produced is uniformly diffused throughout the whole 
 mass of flux, &c., not allowing a particle of silver to escape 
 its action. 
 
 The reducing agent employed in nearly all assays is 
 charcoal, either in its ordinary state, or as it is found in 
 black flux. Starch and other analogous substances may 
 be, as before mentioned, substituted for it : crude argol is, 
 however, the best reducing agent. The portion employed 
 must be varied according to circumstances, so that the 
 silver -lead produced be not too rich, or that too great a 
 proportion of lead be reduced. If the silver-lead be too 
 rich, much of the precious metal may be lost in the slag, 
 and if too great a quantity of lead be produced, silver is 
 again lost, owing to the long exposure to the fire during 
 
602 THE ASSAY OF SILVER. 
 
 cupellation ; and, indeed, this is the most fruitful cause of 
 loss, for more is lost in this manner than by having too 
 little lead produced. In order to know the right propor- 
 tions, the following data will serve as a guide : 1 part of 
 charcoal reduces about 30 parts of lead from litharge, and 
 
 1 part of black flux reduces about 1 part of lead. 
 
 The fluxes employed in this kind of assay are litharge, 
 black flux, potassium or sodium carbonate, and borax. 
 Litharge is an exceedingly convenient flux, because it 
 occupies very little room, and fuses without bubbling, pro- 
 ducing very liquid scoriae with nearly every substance. 
 Experiment has shown that nearly all argillaceous, stony, 
 and ferruginous substances fuse very well with from 8 to 12 
 or more parts of litharge. If from ^ to 1 part of black 
 flux, or 5^- to gV of charcoal, be added to 1 of ore, from 
 J a part to 1 part of silver-lead will be produced. 
 
 Black flux is employed in the fusion of all substances 
 containing a large proportion of alumina, or in which lime 
 is the predominant substance from 2 to 3 parts of this 
 flux generally suffice : 1 part of litharge is added to the 
 assay, which is wholly reduced, producing nothing but 
 lead. 
 
 Potassium or sodium carbonates produce exactly the 
 same effects as the alkali of the black flux. A certain 
 quantity of charcoal must, in this case, be added to the 
 assay. 
 
 Schlutter fuses the poor refuses of goldsmiths' work- 
 shops, mixtures of fragments of crucibles, glass, &c., with 
 
 2 parts of potassium carbonate, when they are very earthy, 
 and with 1 part only when they contain much glass, adding, 
 at the same time, to the mixture a little litharge and 
 granulated lead. 
 
 Borax has, like litharge, the advantage of being a 
 universal flux ; it is useful especially for the fusion of sub- 
 stances containing much lime ; but it is necessary to take 
 great care in the assay, in order to avoid the loss which its 
 boiling up might occasion. This only applies, however, to 
 its use in its ordinary state ; if previously fused that is, 
 used as glass of borax no particular care need be taken. 
 
FUSION WITH OXIDISING REAGENTS. 60S 
 
 FUSION WITH OXIDISING EEAGENTS. 
 
 Litharge. The oxidising agents employed in the assay 
 of argentiferous substances are litharge and nitre. Litharge 
 attacks all the sulphides, arsenio-sulphides, &c., and 
 oxidises nearly all the elements, excepting silver, when 
 employed in sufficient quantity ; and a quantity of lead 
 equivalent to the oxidisable matters present is reduced, so 
 that there results from the assay a slag containing an 
 excess of lead oxide, and an alloy of lead and silver, very 
 little contaminated with foreign metals, if no copper be 
 present, and which can be submitted directly to cupella- 
 tion. This method of assay is exceedingly convenient and 
 quick. 
 
 The pulverised mineral is well mixed with litharge 
 and the mixture placed in a crucible, which may be very 
 nearly filled, as there is scarcely any boiling up when the 
 pot and its contents are submitted to the fire. A thin 
 layer of pure litharge is placed above the mixture, the 
 whole is then heated rapidly, and as soon as the litharge, 
 &c., is completely fused, the crucible is taken from the 
 fire. It is inconvenient to heat it for any length of time, 
 on account of the corrosive action litharge has on the 
 substance of the crucible, which it rapidly destroys. 
 
 The proportion of litharge which must be employed 
 depends upon the nature and quantity of oxidisable 
 matters present in the ore. It ought in general to be very 
 great, because it is absolutely necessary that no sulphurous 
 matters be present, so that the slag may not contain the 
 least trace of silver. But it is known how much litharge 
 is required to decompose the metallic sulphide. Pyrites 
 requires about 50 parts ; mispickel, blende, antimony sul- 
 phide, copper pyrites, grey cobalt, and grey copper require 
 from about twenty-five to about forty times their weight. 
 For sulphide of bismuth 10 are sufficient, and for galena 
 or silver sulphide but 4 or 5 parts need be employed. 
 The proportion of litharge will not be so great for a 
 mineral containing much stony gangue as for one entirely 
 
004 THE ASSAY OF SILVER. 
 
 metallic. Experiment has proved that the assay of rough 
 schlichs, such as those treated in the large way by amalga- 
 mation, can be made very exactly with from 10 to 12 parts 
 of litharge. 
 
 Alloys of silver with the very oxidisable metals, such 
 as those of iron, antimony, tin, zinc, &c., can be assayed 
 by means of litharge ; but in order to have a successful 
 result the alloys should be reduced to a very fine state of 
 division, so that they must be at least granulated ; and it is 
 very often necessary to repeat the operation several, times 
 on the fresh alloy of lead produced. 
 
 The method of assay just pointed out is inconvenient, 
 on account of the large quantity of lead it produces ; pyrites 
 giving 8-J parts, copper pyrites and blende 7 parts, anti- 
 mony sulphide and grey copper about 6 parts, &c. In 
 order to avoid this inconvenience, part of the oxidation 
 can be performed by means of nitre. Nitre alone, em- 
 ployed in excess, oxidises all metallic and combustible 
 substances found with silver, and even, under certain cir- 
 cumstances, a portion of the silver itself ; but when the 
 proportion is insufficient to oxidise the whole, and when 
 the mixture contains at the same time litharge, after the 
 nitre has produced its action the litharge acts in its turn 
 on the remainder of the oxidisable substances, and the 
 resulting lead carries down the silver set free. So that, by 
 employing suitable proportions of nitre and litharge, all the 
 silver contained in oxidisable minerals may be extracted, 
 and any quantity of lead required may be thus alloyed 
 with it. 
 
 As to the requisite proportion of nitre, it can be come 
 at by practice, aided by the following data. It requires 
 about 2^ parts of nitre to completely oxidise iron pyrites, 
 1^ for sulphide of antimony, and f for galena. 
 
 This estimation can be ascertained at once as fol- 
 lows : Fuse 1 part of the mineral with 30 of litharge, and 
 weigh the resulting button of lead ; and having fixed upon 
 the quantity of lead necessary to carry on the cupellation 
 properly, deduct it from the whole weight of the button, 
 and the difference will be the amount of lead necessary to 
 
SPECIAL DIRECTIONS FOR THE CRUCIBLE ASSAY. 605 
 
 leave in the slag in the state of oxide ; and as it has been 
 proved by experiment that 1 part of lead requires -25 to 
 30 of nitrethat is, from 25 to 30 per cent. it is easy 
 to calculate the quantity necessary to be added. 
 
 When the ore contains sulphur, the latter forms with 
 the nitre, potassium sulphate, which swims on the slag with- 
 out combining with it. 
 
 The assay of silver ores by means of nitre is advan- 
 tageous and useful in a variety of cases. If we wish to esti- 
 mate, for example, very exactly the percentage of silver 
 in a poor galena, a large quantity, say a quarter of a pound, 
 must be fused with about an ounce or an ounce and a half 
 of nitre, and a quarter of a pound of sodium carbonate, 
 or, better still, the same quantity of litharge, one of either 
 of which must be employed to flux the gangue and tem- 
 per the deflagration. After the fusion, all the contained 
 silver will be found alloyed with a very small quantity of 
 lead. 
 
 Sometimes the assay is made with a larger quantity of 
 nitre than is requisite for the oxidation, and when the 
 mixture is perfectly fused a certain quantity of metallic 
 lead is added, taking care to cover the whole surface of 
 the mixture, either by using granulated lead or a conve- 
 nient mixture of litharge and charcoal, or litharge and 
 galena. The shower of metallic lead passing through the 
 fluid mass alloys with all the silver it finds in its passage, 
 and so concentrates it. 
 
 This process, however, cannot always be confidently 
 employed. If an excess of nitre be employed with sub- 
 stances susceptible of forming peroxides capable of attack- 
 ing silver, such as some cupreous substances, the lead added 
 reduces the greater part, but not the whole of the silver in 
 the ore, so that the assay will not be perfect. 
 
 Special Directions for the Crucible Assay of Ores and 
 Substances of the First Class. 
 
 The ores and substances belonging to this class may, 
 for the convenience of assay, be further subdivided on the 
 
606 THE ASSAY OP SILVER. 
 
 following principle. It has already been seen that sulphur, 
 and other substances having a great affinity for oxygen, 
 reduce metallic lead from litharge in proportion to the 
 amount of reducing matter present ; and as it is necessary 
 in this kind of assay that no more than a certain quantity 
 of lead alloy should be submitted to cupellation, some kind 
 of control must be exercised by thje assayer, to keep the 
 quantity of lead reduced in due and proper bounds. This 
 is readily accomplished by what is called a ' preliminary 
 assay,' by which all ores and substances of this class are 
 divided into three sections : 1st, ores which, on fusion 
 with excess of litharge, give no metallic lead, or less than 
 their own weight ; 2ndly, those which give their own 
 weight, or nearly their own weight, of metallic lead ; 3rdly, 
 those which give more than their own weight of metallic 
 lead. The preliminary or classification assay is thus con- 
 ducted : 
 
 Carefully mix 20 grains of the finely pulverised ore 
 (all silver ores must be passed through a sieve with 80 
 meshes to the linear inch) with 500 grains of litharge ; 
 place the mixture in a crucible which it only half fills ; 
 set the crucible, after careful warming, in a perfectly bright 
 fire, and get up the heat as rapidly as possible, so as to 
 finish the operation in a short time, to prevent the action 
 of the reducing gases of the furnace on the lead oxide ; 
 because if a great length of time were taken in the opera- 
 tion, a portion of the lead reduced might be traceable to 
 the furnace gases, and the result of the experiment vitiated. 
 After the contents of the crucible are fully fused, and the 
 surface perfectly smooth, the crucible may be removed 
 and allowed to cool, and when cold broken. One of 
 three circumstances may now present itself to the assayer : 
 1st, no lead, or less than 20 grains, has been reduced ; 
 2ndly, 20 or nearly 20 grains, more or less, may be 
 reduced ; and Srdly, more than 20 grains may have been 
 reduced. 
 
 Now, as it has been already stated, 200 grains of lead 
 alloy is a suitable amount to cupel, and as 200 grains is 
 the best quantity of ore to submit to assay, it will be 
 
SPECIAL DIRECTIONS FOR THE CRUCIBLE ASSAY. 607 
 
 evident that ores and substances of the second section, or 
 those bodies which give their own weight, or nearly their 
 own weight, of lead alloy, simply require fusion with a suit- 
 able quantity of litharge and an appropriate flux. Ores of 
 the first section require the addition of a reducing agent, in 
 quantity equivalent to the standard amount of lead alloy 
 (200 grains) ; and ores of the third section require an equi- 
 valent quantity of an oxidising agent, or an amount of some 
 substance which will oxidise the lead in excess of 200 
 grains of alloy. 
 
 The reducing agent employed is argol ; the oxidising 
 agent potassium nitrate. It is necessary, before commenc- 
 ing an assay of a silver ore, to estimate how much lead a 
 given weight of the argol the assayer has in use will reduce ; 
 as also how much lead a given weight of potassium nitrate 
 will oxidise. These assays are thus made : 
 
 Assay of Reducing Power of ArgoL Carefully mix 20 
 grains of the argol to be tested with 500 grains of litharge 
 and 200 grains of sodium carbonate ; place the mixture in 
 a suitable crucible, and cover with 200 grains of common 
 salt. (It is best to mix two such quantities, and take the 
 mean of the results.) Fuse with the precautions pointed 
 out in assay of substances of the first class, containing 
 lead. 
 
 Weigh the resulting buttons, and take a note of the 
 mean weight, which will represent the amount of lead 
 reducible by 20 grains of argol. 
 
 Assay of Oxidising Power of Potassium Nitrate. Mix 
 20 grains of finely powdered potassium nitrate, 50 grains of 
 argol, 500 grains of litharge, and 200 grains of sodium 
 carbonate ; cover with 200 grains of common salt, and fuse 
 as above. Weigh the resulting button. Now calculate 
 the amount of lead which should have been reduced by 50 
 grains of argol, and the difference between that and the 
 amount of lead reduced in this experiment will represent 
 the amount of lead oxidised by 20 grains of potassium 
 nitrate. 
 
 Thirty to 32 grains of ordinary red argol reduce about 
 200 grains of lead ; and 23 grains of pure potassium nitrate 
 
(508 THE ASSAY OF SILVEE. 
 
 oxidise about 100 grains of lead. The assayer must, 
 however, adopt the numbers found by himself by experi- 
 ment, as the samples of argol and nitre may be more or 
 less impure. He must also examine every fresh supply of 
 litharge for the amount of silver it contains, in the follow- 
 
 ing manner : 
 
 Assay of Litharge for Silver. Mix 1,000 grains of 
 litharge with 30 grains (or any other quantity that may, 
 by experiment, be found requisite) of argol, 200 grains of 
 sodium carbonate, and cover with salt, as already directed. 
 Fuse the mixture in a suitable crucible ; allow it to cool ; 
 break and cupel the button obtained, as hereafter to be 
 described ; take a note of the amount of silver obtained ; 
 and as 1,000 grains of litharge is the standard quantity for 
 a silver assay, the amount of silver, indicated as above, is 
 to be deducted from the amount of silver obtained in the 
 assay of any silver ore, until that quantity of litharge is 
 consumed. 
 
 Assay of Ores of the First Section. Make a preliminary 
 assay, as already described. Suppose 10 grains of lead 
 result ; then, as 20 have furnished 10 grains, so 200 grains 
 of ore would furnish 100 grains of lead, or 100 grains less 
 than the quantity best adapted for cupellation ; so that, 
 referring to the assay of argol, and finding that from 30 to 
 32 grains reduce 200 grains of lead, then it is clear that 
 the reducing power of from 15 to 16 grains of argol, in 
 addition to the reducing power of 200 grains of ore, is 
 necessary to furnish 200 grains of lead alloy. In this case 
 the ingredients required in the actual assay, or ' assay 
 proper,' would stand thus : 
 
 200 grains of ore. 
 200 grains of sodium carbonate. 
 1,000 grains of litharge. 
 15 to 16 grains of argol. 
 
 These materials are to be thoroughly well mixed, 
 placed in a crucible which they about half fill, and covered 
 first with 200 grains of common salt, and then 200 grains 
 of borax, and submitted to the fire with the usual precau- 
 
ASSAY OF ORES OF THE THIRD SECTION. 609 
 
 tion ; when the flux flows smoothly the assay is complete ; 
 it may be removed and allowed to cool, the crucible broken, 
 and the button obtained must be hammered into a cubical 
 form, and should approximate to 200 grains, either more 
 or less, within 10 grains. Two crucibles must always 
 be prepared. It will also be here convenient to mention 
 that the argol and potassium nitrate are the only sub- 
 stances whose quantities vary in the assay of silver ores, 
 the amount of these variations being estimated by the 
 preliminary or classification assay. 
 
 Assay of Ores of the Second Section. If the preliminary 
 assays of the sample submitted to assay furnish from 18 
 to 22 grains of lead, then the assay proper may be thus 
 made : 
 
 200 grains of the ore, 
 200 grains of sodium carbonate, 
 1,000 grains of litharge, 
 
 well mixed and covered with salt and borax as above. 
 Fuse with due care, and reserve buttons of lead alloy for 
 cupellation. 
 
 Assay of Ores of the Third Section. If the sample on 
 preliminary assay furnished 40 grains of lead, then the 200 
 grains employed in assay proper would give 400 grains or 
 200 grains of lead in excess ; refer now to note-book for 
 quantity of lead oxidised by nitre ; suppose the nitre pure 
 as just stated, 23 grains will oxidise 100, therefore 46 
 grains are equivalent to 200, and the assay proper will 
 stand thus : 
 
 200 grains of the ore. 
 
 200 grains of sodium carbonate* 
 1,000 grains of litharge. 
 
 46 grains of potassium nitrate. 
 
 The potassium nitrate is to be weighed first, finely pul- 
 verised, and then well mixed with the remaining sub- 
 stances, and covered with salt and borax. The crucible 
 in this assay must be larger than in the two preceding 
 cases ; the mixture should not more than one-third fill it, 
 
 R R 
 
010 THE ASSAY OF SILVEE. 
 
 as there is a considerable action set up between the oxygen, 
 of the nitre and the sulphur or arsenic, or any other sub- 
 stance that may be the reducing agent in the ore ; for in 
 fact the nitre does not directly oxidise the lead, which 
 sulphur, &c., might have reduced, but oxidises its equiva- 
 lent quantity of sulphur, or whatever other reducing 
 substance there may be in the ore, so as only to leave a 
 sufficient amount to reduce 200 grains of lead, in lieu of 
 the 400 as indicated by preliminary assay, or when the 
 reducing power of the ore was allowed to come into full 
 play. The buttons obtained in this case are also to be 
 reserved for cupellation. 
 
 Scarification. Scorification has, like fusion with lith- 
 arge, the effect of producing an alloy of lead capable of 
 cupellation, and a very fusible slag composed of lead oxide, 
 and all the matters foreign to silver, converted into the 
 state of oxide. In the crucible assay as just described the 
 oxidation of these substances takes place by the action 
 of the litharge, which furnishes at the same time by its 
 reduction the lead necessary to form the alloy, whilst in 
 scorification all the substances susceptible of oxidation 
 are oxidised in the roasting by means of the oxygen of 
 the air, and the litharge itself is produced by the oxidation 
 of part of the lead mixed with the ore to be assayed. 
 
 In this operation vessels termed scorifiers (see p. 144) 
 are employed. They are heated in the muffle of the 
 cupelling furnace, and as many assays may be made at 
 one time as the muffle holds scorifiers. 
 
 Before introducing the scorifiers into the muffle, a given 
 weight of the ore reduced to powder is mixed intimately 
 with a certain quantity of granulated lead, and placed in 
 each. They must then be heated gradually for about a 
 quarter of an hour, with the door of the muffle closed, in 
 order to fuse the lead ; then diminish the heat and allow 
 access of air by opening the door. The current thus esta- 
 blished in the muffle soon causes the commencement of the 
 roasting ; and this roasting goes on without its being neces- 
 sary to continually agitate the mass, as in the case of pul- 
 verulent substances. 
 
SCOKIFICATION. 611 
 
 During the oxidation, a slag is formed on the fluid 
 metal, which is thrown towards the edges, and which, by 
 continually augmenting, at last entirely covers the bath. 
 This slag, which is often solid at the commencement, be- 
 comes softer and softer, and at last becomes perfectly fluid ; 
 because, in proportion to the advance of the operation, the 
 proportion of lead oxide continually increases. When it 
 is judged that the scorification has been carried far enough, 
 the melted matter is stirred with a rod of iron, in order to 
 mix with the mass the hard or pasty parts attached to the 
 bottom or sides of the scorifier. The fire is then urged so 
 as to completely liquefy the slags. It may be ascertained 
 when they are sufficiently fluid by plunging into them a 
 red-hot iron rod, which must only be covered with a slight 
 coating, capable of running off, and not solidifying into a 
 drop at the end. 
 
 This condition of liquidity is indispensable, in order to 
 enable the metallic globules to unite into a single button. 
 When this end is not attained, it is because the scorifica- 
 tion has not been carried sufficiently far, or because a 
 sufficient quantity of lead has not been added to form the 
 flux, in which case a fresh quantity must be added, or, 
 what is preferable, the assay recommenced with larger 
 proportions. 
 
 When the operation is finished, the scorifier must be 
 removed, and its contents immediately poured into a 
 circular or hemispherical ingot mould (see fig. 29, p. 69). 
 The metallic particles fall to the bottom, and as the cooling 
 proceeds they form ' a button covered by the slag, which 
 is readily detachable by a blow of a hammer ; it ought 
 to be very homogeneous and vitreous, and its colour vary- 
 ing from brown to greenish. 
 
 It is always advisable to examine the slag, and ascertain 
 if it contain metallic globules. The button ought to be as 
 ductile as ordinary lead ; if not, it cannot be cupelled, and 
 must be submitted to a fresh operation. It is in general 
 advantageous to push the scorification to its greatest ex- 
 tent, because experiment has proved that less silver is lost 
 than when a large button is cupelled. Nevertheless, there 
 
 R R 2 
 
612 THE ASSAY OF SILVER. 
 
 is a limit, because if the silver-lead produced be too rich, 
 the least loss in the shape of globules would cause a not- 
 able one in the silver. Besides, as litharge exercises a very 
 corrosive action on earthy matters, if the scorification be 
 continued for a great length of time, it sometimes happens 
 the vessel is pierced, and the assay has to be recommenced. 
 The button of lead remaining ought to weigh about 200 to 
 300 grains, when the ores treated are of ordinary richness. 
 The length of time a scorification takes is from half an 
 hour to an hour. The scorifier can be rendered less per- 
 meable to the litharge by being rubbed inside with chalk, 
 or, better still, red ochre. 
 
 There may be distinguished three distinct periods in the 
 operation viz. the roasting, the fusion, and the scorifica- 
 tion. At first a strong fire is employed ; but the doors ol 
 the furnace are opened as soon as the mixture is fused. 
 The mineral, being specifically lighter than the lead, is then 
 seen floating on its surface, or forming masses in it ; the 
 roasting then commences, and from the appearance of the 
 vapours the nature of the combustible matter it contains 
 may be judged. Sulphur produces clear grey vapours ; 
 zinc, blackish vapours, and a brilliant white flame ; arsenic, 
 whitish-grey vapours ; antimony, fine red vapours, &c. 
 When no more fumes are seen, the mineral has disap- 
 peared, and the fused lead is perfectly uncovered, the 
 roasting has terminated ; this generally requires from 
 eighteen to twenty minutes. At this time the fire is 
 urged, so as to cause all the substances in the scorifier to 
 fuse. It can be ascertained that the fusion is complete by 
 the following signs : at the instant the muffle is opened the 
 button becomes whitish-red with a greyish- black band, 
 and there arise from the melted mass clear white fumes of 
 lead, and the slag appears like a ring encircling the metal. 
 The third period then commences : the furnace is cooled, 
 as in the roasting, and the lead is allowed to scorify until 
 it is entirely covered with fused oxide ; this last period 
 generally lasts about fifteen minutes. The fire is then 
 increased for about five minutes, and the contents of the 
 scorifier poured into the mould. 
 
 
SCORIFICATIOtf* 613 
 
 The process of scorification is applicable to all argent- 
 iferous matters, and is at the same time the most exact 
 method of assay, as also the most convenient, when a large 
 number of assays are required at the same time, because 
 they are entirely executed in the muffle, which, with most 
 assay ers, is generally hot : it, however, requires a greater 
 number of vessels as cupels, &c. 
 
 When the silver ores are stony, the lead oxide formed 
 during the roasting combines with the gangue, forming a 
 fusible compound, whilst the remaining lead alloys with 
 the silver. When the ores are metallic, the oxidisable 
 bodies absorb oxygen from the atmosphere ; and the 
 oxides so formed combine with the litharge produced at 
 the same time, forming a compound which becomes very 
 fusible in proportion as the lead oxide increases ; and if 
 the scorification has not been pushed sufficiently far, the 
 button will contain, besides silver and lead, a little copper, 
 which will not, however, interfere with the cupellation. 
 There is this one peculiarity about scorification, that, how- 
 ever small the proportion of lead may be that is used, at 
 the end of the operation the slag does not contain any 
 oxysulphide. For instance, even when oxysulphides are 
 produced in the course of scorification, they are completely 
 decomposed in the roasting, and in consequence it is very 
 rarely that the slag retains any proportion of silver ; and 
 as to the proportion of lead employed, only just enough 
 to render the slag liquid, and to produce sufficient lead 
 for cupellation, is necessary. 
 
 It is different, however, when the sulphides and arsenic- 
 sulphides are assayed by means of litharge ; for from 30 
 to 50 parts of that substance must be employed to prevent 
 the scoriae retaining any silver, or, as already pointed out, 
 a certain proportion of nitre must be added. 
 
 All scorifications may be conducted by the simple 
 addition of lead ; but it has been proved that the opera- 
 tion proceeds more quickly, and with less danger to the 
 scorifier, when borax is employed. This salt dissolves the 
 oxides in proportion as they are produced, as also the 
 gangues, and forms a very liquid slag from the commence- 
 
614 THE ASSAY OF SILVER. 
 
 ment of the operation, which does not happen when lead 
 alone is used, because litharge, which can alone cause the 
 fusion, is only present in the slag in sufficient proportion 
 at a very advanced stage of the operation. 
 
 When the slag is liquid at the beginning of the opera- 
 tion (as occurs in the use of borax), it is continually thrown 
 on the sides of the scorifier, and forms a ring on the sur- 
 face of the bath, leaving in the centre the metallic sub- 
 stance, having a considerable extent of surface which is 
 continually diminishing. 
 
 The current of air, being thus directly in contact with 
 the fused metals, rapidly causes their oxidation, which 
 does not take place when the semi-fluid substances float 
 here and there on the metallic bath. The proportion of 
 lead and borax necessary for a scorification varies exceed- 
 ingly, according to the nature of the substance under 
 assay, and ought to be greater in proportion as the sub- 
 stances, or resulting oxides, are difficult of fusion. In 
 ordinary cases 12 parts of lead and 1 of glass of borax, 
 are employed ; but sometimes 32 of lead, and 3 of borax, 
 are required. A large proportion of borax is useful, 
 especially when the substances contain much lime, zinc 
 oxide, or tin oxide. 
 
 Instead of borax, glass of lead may be employed. It 
 acts as a flux on silica ; but its action is much less effec- 
 tive than that of borax. 
 
 There are some substances which scorify with a small 
 proportion of lead. Thus, for galena and copper sulphide, 
 2 parts of lead suffice ; but 8 parts are required for ores 
 which contain much gangue. 
 
 Silver antimonide can be scorified with 8 parts of lead, 
 but according to experiments made in the Hartz it appears 
 that the slag retains about yy^th of silver ; with 16 parts 
 'of lead 2-^-0 th of fine metal is still lost ; but with 3 of borax 
 and 16 of lead not the slightest trace remains in the slag. 
 t It is very difficult to separate tin and silver by the dry 
 way. The best method is to roast the alloy in a scorifier, 
 adding to it 16 parts of lead and 3 of borax at least, and 
 operating as before described. 
 
ASSAY OP SUBSTANCES OF THE FIRST CLASS. 615 
 
 Speiss almost always contains silver, and is one of the 
 most difficult substances to assay. If nickel be present, 
 the button cannot be cupelled. Generally, speiss may be 
 scorified with 16 parts of lead ; and the same operation is 
 gone through twice or thrice, adding each time a fresh 
 quantity of lead. The operation would probably succeed 
 by roasting the speiss in the scorifier before adding the . / 
 lead. 
 
 Special Instructions for the Scorification Assay of Ores 
 of the First Class. This mode of assay has an advantage 
 over the crucible assay just described, inasmuch as if 
 properly conducted no preliminary assay is required ; but 
 this is greatly counterbalanced by the fact that not more 
 than 50 grains of ore can be operated on in one scorifier, 
 and that good or trustworthy results cannot be obtained 
 by this method unless four scorifier s are employed for 
 each assay, so that in all 200 grains of ore may be em- 
 ployed. There are thus employed four scorifiers to three 
 crucibles, and four cupels to two cupels ; as in one case 
 four buttons are to be submitted to cupellation, and in 
 the other only two. When very rich copper ores, how- 
 ever, have to be assayed for silver, the plan by scorifica- 
 tion is very useful, as in the crucible operation much 
 copper is reduced with the lead, so as to require a very 
 large quantity of lead for its conveyance as oxide into the 
 cupel. This class of assay will, however, be particularly 
 noticed under the head Assay of the Alloys of Silver. 
 
 Assay in Scorifier. Weigh out 300 grains of granu- 
 lated lead, place them in a scorifier, then add 50 grains 
 of pulverised fused borax, and 50 grains of the ore to be 
 assayed, well mix them in the scorifier by aid of a spatula, 
 and cover the mixture with other 300 grains of granu- 
 lated lead : prepare in this way four scorifiers, place them 
 in the muffle with the tongs (b, fig. 28, page 68), and care- 
 fully watch them with all the precautions before pointed 
 out : when the surface of the metal is quite covered with 
 fused oxide, pour the contents of each scorifier into one 
 of the hollows of the mould depicted at fig. 29, page 69. 
 When the mass of slag and metal is cold, separate the 
 
616 THE ASSAY OF SILVEE. 
 
 latter from the former by means of the hammer and anvil, 
 hammer the metal into the form of a cube, and reserve it 
 for cupellation. 
 
 Assay of Substances of the First Class admixed with 
 Native or Metallic Silver. The same kind of calculation 
 is necessary in the assay of ores as above, as in the case of 
 copper ores containing metallic copper. The sample must 
 be carefully weighed. Suppose it to weigh 2,500 grains. 
 It must be pulverised, and as much as possible passed 
 through the sieve with eighty meshes to the linear inch. 
 It will be thus divided into two parts: the one passing 
 through the sieve is mineralised silver that is, silver ore 
 of various kinds mixed with earthy matter, and a very 
 small quantity of metallic silver which has been sufficiently 
 divided to pass through a sieve of such a degree of fine- 
 ness ; the other, impure metallic silver, which has been 
 unable to pass through the sieve. The weights of both 
 portions are carefully taken, and thus noted 
 
 . Kough metallic silver . . . 5'07 grs. 
 
 Ore through sieve .... 2494*93 
 Total weight of sample . . . 2500-00 
 
 Assay the ore which passed through the sieve as 
 already directed, and the rough silver as directed under 
 the head Assay of Silver Alloys. Note the quantity of 
 silver obtained in each experiment. Thus : suppose 200 
 grains of ore yielded 2 grains of fine silver, and the 5 -07 
 grains of rough silver 4 grains of fine silver by cupellation, 
 the number of ounces of fine silver in the ton is thus 
 calculated. 
 
 On referring to Table III. in Appendix, it will be found 
 that if 200 grains of ore yield 2 grains of fine silver ,. 
 1 ton will yield 326 ozs. 13 dwts. 8 grs. of fine silver; 
 so that the average produce of the ore is the above 
 amount. 
 
 Then, if 5 -07 grains of rough silver yield 4 grains of 
 fine silver, 200 grains would yield, by calculation, 159;763 
 grains of fine silver. 
 
 Thus 
 
ASSAY OP SJJBSTANCES OF THE FIRST CLASS. 617 
 
 Now, by referring to Table III. in the Appendix, it 
 will be found that 200 grains of ore give 159 grains of fine 
 silver=25,970 ounces per ton; and that 200 grains of 
 ore give '763 grain of fine silver =124 ozs. 12 dwts. 
 11 grains : therefore, the 5 -07 grains of rough silver con- 
 tain at the rate of 26,094 ozs. 12 dwts. 11 grs. per ton, 
 thus 
 
 25,970 ozs. + 124 ozs. 12 dwts. 11 grs. = 26,094 ozs. 12 dwts. 11 grs. 
 
 Thus we have 
 
 ozs. dwts. grs. 
 
 Average produce of ore .... 326 8 
 
 Average produce of rough silver . . 26,094 12 11 
 
 per ton of 20 cwts. 
 
 Then, as in the case of the copper, multiply the weight 
 and produce of each portion together, add the resulting 
 total products, and divide the sum by the weight of the 
 sample. For this purpose it is better to reduce the penny- 
 weights and grains to their decimal values. Thus 13 dwts. 
 8 grs. is nearly equal to *67 of an ounce, and 12 dwts. 11 
 grs. to -62 of an ounce ; therefore the quantities above will 
 stand thus, 326-67 ozs. and 26,094-62 ozs. 
 
 Then 326-67 x 2494-93 = 815018-7831 
 >and 26094-62 x 507 = 132296-7234 
 and 815018-7831 + 132299-7234 
 
 2500 
 
 = 378-9 oz. 
 
 or 378 ozs. 18 dwts. (nearly) per ton of the original sample,, 
 before pulverising and sifting. 
 
 In every case of assay yet described, it may be men- 
 tioned that if the sample contain gold, the whole of that 
 metal will be found with the silver, as obtained by cupel- 
 lation, and may be separated as stated in the chapter on 
 the Assay of Gold. 
 
 Cupellation. Cupellation is an operation that has been 
 known from time immemorial ; it has many characters in 
 common with scorification, and is effected in nearly the 
 same manner. Like that, it has for its end the separation 
 of silver and gold from different foreign substances, by 
 means of lead ; but it differs in this, that the scoriae pro- 
 duced are absorbed by the substance of the vessel named 
 
618 THE ASSAY OF SILVER. 
 
 a cupel, in which the operation is made, instead of remain- 
 ing on the melted metal, the latter remaining uncovered 
 and in contact with the air, so that the extraneous metals 
 are not only oxidised, but also all the lead ; and there re- 
 mains nothing but the pure metals, silver and gold, or an 
 alloy of them, in the cupel. 
 
 Cupellation requires, as an indispensable condition, 
 that the slag should have the property of penetrating and 
 soaking into the body of the substance forming the cupel ; 
 it is, therefore, only applicable to a certain number of 
 substances, and not to all, like scorification. Lead and 
 bismuth oxides, in a state of purity, are the only oxides 
 which possess the property of soaking into the cupel ; but 
 by the aid of one or the other, various oxides which by 
 themselves form infusible scorise on the cupel, acquire the 
 property of passing through it : therefore, on making a 
 cupellation, it is necessary to fuse the substance with a 
 sufficient proportion of lead or bismuth, so that the oxides 
 they produce may combine with the oxides of all the foreign 
 metals produced in the operation, and carry them into the 
 body of the cupel. 
 
 This proportion varies with the nature of the sub- 
 stances cupelled, and other circumstances. The quantity 
 required in ordinary cases will be mentioned hereafter. 
 
 The cupels or porous vessels in which the operation is 
 made ought to have a sufficiently loose texture to allow 
 the fused oxides to penetrate them easily, and at the same 
 time to possess sufficient solidity to enable them to bear 
 handling without fracture ; and, moreover, they ought to 
 be of such a nature as not to enter into fusion with either 
 lead or bismuth oxide. For a description of their mode of 
 manufacture, see p. 141. 
 
 The following is the method in which an ordinary 
 -cupellation is conducted : The furnace being heated, the 
 bottom of the muffle is covered with cupels, placing the 
 largest towards the end ; and if they are required to be 
 heated as quickly as possible, they may be placed upside 
 down, and turned, at the instant of use, by means of the 
 tongs. When the interior of the muffle is reddish-white, 
 
ASSAY OF SUBSTANCES OF THE FIRST CLASS. 610 
 
 the matters to be cupelled may be introduced. When the 
 cupels have been placed in their proper position, great 
 care must be taken from the commencement to blow out 
 of them all cinders, ashes, and other extraneous substances 
 which may have fallen into them. 
 
 The substance to be cupelled is sometimes an alloy, 
 which can pass without addition of lead, and sometimes 
 a compound, to which lead must be added. In the first 
 case, the alloy is taken hold of by a small pair of forceps, 
 and deposited gently in the cupel. In the second case, the 
 substance to be cupelled is enveloped in a sheet of lead of 
 suitable weight, and placed, as before, in the cupel ; or the 
 necessary quantity of lead may be first placed in the cupel, 
 and when the lead is fused, the substance to be cupelled 
 added, taking care not to agitate the melted mass and 
 cause loss by splashing. If the substance to be cupelled 
 is in very small pieces, as grains or powder, it must be 
 enveloped in a small piece of blotting-paper, or, still 
 better, in a piece of very thin sheet lead, giving it a slightly 
 spherical form, and dropping it gently into the mass ot 
 molten metal in the cupel. Sometimes the substance is 
 gradually added, by means of a small iron spoon ; but it 
 is preferable to use paper or thin lead, as just recom- 
 mended. 
 
 When the cupels are filled, the furnace is closed, either 
 by the door or by pieces of lighted fuel, so that the fused 
 metals may become of the same temperature as the muffle. 
 When this point has been gained, air is allowed to pass 
 into the furnace ; the metallic bath is then in the state 
 termed uncovered ; that is, it presents a convex surface, 
 very smooth and without slag. When the air comes in 
 contact with it, it becomes very lustrous, and is covered 
 with luminous and iridescent patches, which move on the 
 surface, and are thrown towards the sides. These spots 
 are occasioned by the fused oxide of lead which is continu- 
 ally forming, and which, covering the bath with a very 
 thin coating of variable thickness, presents the phenome- 
 non of coloured rings. 
 
 The fused litharge, possessing the power of moistening 
 
020 . THE ASSAY OF SILVER 
 
 (so to speak) the cupel, is rapidly absorbed by it when 
 sufficiently porous, so that the metallic alloy is covered 
 and uncovered every instant, which establishes on its sur- 
 face a continual motion from the centre to the circumfer- 
 ence. At the same time a vapour rises from the cupels 
 which fills the muffle, and is produced by the vapour of 
 lead burning in the atmosphere. An annular spot is soon 
 observed on the cupel around the metal, and this spot 
 increases incessantly until it has reached its edges. 
 
 In proportion as the operation proceeds, the metallic 
 bath of silver-lead diminishes, becoming more and more 
 rounded ; the shining points with which it is covered 
 become larger and move more rapidly ; lastly, as the 
 whole of the lead separates, the button seems agitated by 
 a rapid movement, by which it is made to turn on its 
 axis ; it becomes very lustrous, and presents over its 
 whole surface all the tints of the rainbow : suddenly the 
 agitation ceases, the button becomes dull and immovable, 
 and after a few instants it takes the look of pure silver. 
 This last part in the operation of cupellation is termed the 
 brightening, fulguration, or coruscation. 
 
 If the button be taken from the muffle directly after 
 the brightening, it may throw off portions of its substance; 
 this must be avoided, especially when the button is large. 
 The button, when covered by mammillated and crystalline 
 asperities, is said to have ' vegetated.' The cause of this 
 effect seems to be, that when the fused buttons are suddenly 
 exposed to the cold air, the silver solidifies on the surface, 
 whilst that in the interior remains liquid. The solid crust, 
 contracted by cooling, strongly compresses the liquid 
 interior, which opens passages for itself, through which it 
 passes out, and around which it solidifies when in contact 
 with the cool air. But it sometimes happens that, when 
 the contraction is very strong, a small portion of the silver 
 is thrown off in the shape of grains, which are lost. 
 
 After brightening, the cupels must be left for a few 
 minutes in the furnace, and drawn gradually to the mouth, 
 before they are taken out, so that the cooling may be slow 
 and gradual. These precautions are nearly superfluous 
 
CUPELLATION. 621 
 
 when the buttons are not larger than the head of an ordi- 
 nary pin. 
 
 As silver is sensibly volatile, it is essential, in order 
 that the smallest possible quantity be lost, to make the 
 cupellation at as low a temperature as may be. On the 
 other hand, the heat ought to be sufficiently great, so that 
 the litharge may be well fused and absorbed by the cupel ; 
 and, moreover, if the temperature be too low, the opera- 
 tion lasts a very long time, and the loss by volatilisation 
 will be more considerable than if 'the assay had been made 
 rapidly at a much higher temperature. 
 
 Experience has proved that the heat is too great when 
 the cupels are whitish and the metallic matter they contain 
 can scarcely be seen, and when the fume is scarcely visible 
 and rises rapidly to the arch of the muffle. On the con- 
 trary, the heat is not strong enough when the smoke is 
 thick and heavy, falling in the muffle, and when the 
 litharge can be seen not liquid enough to be absorbed, 
 forming lumps and scales about the assay. When the 
 degree of heat is suitable the cupel is red, and the fused 
 metal very luminous and clear. 
 
 In general, it is good to give a strong heat at the com- 
 mencement, so as to well uncover the bath, then to cool 
 down, and increase the heat at the end of the operation for 
 a few minutes, in order to aid the brightening. There can 
 be no inconvenience in urging the temperature at first, be- 
 cause the silver-lead is then poor, and much precious metal 
 cannot be lost by volatilisation. The increase of fire given 
 towards the end is for the purpose of separating the last 
 traces of lead, from which it is very difficult to free the 
 silver ; but this strong fire must not be continued long, 
 otherwise there might be a notable loss by volatilisation. 
 When the assay of very poor argentiferous matters is 
 made, the heat can be kept up nearly all through the 
 cupellation. It generally succeeds better when the tem- 
 perature is too high than too low. 
 
 The force of the current of air which passes through 
 the muffle is another very important thing in the success 
 of the operation. Too strong a current cools the cupel, 
 
022 THE ASSAY OF SILVER. 
 
 oxidises too rapidly, and the assay would be spoilt. With 
 too feeble a current the operation proceeds slowly, the 
 assay remains a long time in the fire, and much silver is 
 lost by volatilisation. 
 
 When the litharge is produced more rapidly than it 
 can be absorbed by the cupel, or when it is not liquid 
 enough, which may happen from the furnace being too 
 cold, or when other oxides, produced at the same time, 
 diminish its fusibility, it accumulates gradually on the fluid 
 metal, forming at first a ring which envelopes its circum- 
 ference, and which, gradually extending, covers the whole 
 surface ; at this period the assay becomes dull, and all 
 movement ceases. When the operation is carefully at- 
 tended to, it is nearly always possible to avoid this acci- 
 dent. If at the first moment any signs are manifested of 
 this evil, the temperature of the muffle must be raised, 
 either by shutting the door, or placing in it burning fuel ; 
 the assay will, in a little time, resume its ordinary course. 
 But when the cause of the mishap is supposed to be the 
 abundance of foreign oxides in the assay, a fresh propor- 
 tion of lead must be added. 
 
 It can be ascertained whether an assay has passed well 
 by the aspect of the button. It ought to be well rounded, 
 white, and clear, to be crystalline below, and readily de- 
 tached from the cupel. When it retains lead it is brilliant 
 below and livid above, and does not adhere at all to the 
 cupel. 
 
 Ill order to detach the button, seize it with a strong 
 pair of pliers (see fig. 109), and examine with a microscope, 
 (see fig. HO), brushing it to detach small particles of 
 litharge which may adhere to it, and place it in the pan 
 of a balance (fig. 13, page 28), which will indicate the 
 T _i__tn of a grain. The weight of the silver furnished by 
 the lead or litharge employed in the operation ought to 
 be subtracted from the amount of silver obtained ; so that 
 it is necessary to ascertain the richness of these matters 
 beforehand, as they are never completely free from silver. 
 The poorest of them contain from TWoo tn to iw^o- tn - 
 
 Sometimes an equal quantity of lead is placed in another 
 
CUPELLATION. 623 
 
 cupel, and the silver thus obtained placed in the balance- 
 pan containing the weights. 
 
 Cupellation does not give the exact proportion of silver 
 contained in an alloy. There is always a loss, and this loss- 
 is always greater than that which takes place in the large 
 way, as in the latter process a greater quantity is always 
 
 FIG. 109, FIG. 110. 
 
 obtained than that estimated by the assay. The loss of 
 silver is traceable to three causes .: 1st, to volatilisation ; 
 2ndly, to oxidation ; Srdly, and lastly, to the absorption of 
 minute globules of silver into the body of the cupel. It is 
 certain volatilisation takes place, because a notable quan- 
 tity of silver is always found deposited on the sides of the 
 furnace and chimney in the shape of dust ; and silver, which 
 is volatile by itself, becomes much more so when alloyed 
 with lead, and is carried away by the vapours of the latter, 
 and found in the pulverulent deposits, termed lead smoke 
 or fume, which proceed from the combustion of the latter 
 
624 THE ASSAY OF SILVER. 
 
 metal in the air. Nevertheless, this cause of loss is not 
 very important, for it is rare that the fume contains more 
 than -roiro-oth f silver, and accurate experiments have 
 proved that in cupellation in the small way not more than 
 two to three per cent, of lead is volatilised. It is certain 
 that a portion of the silver found in cupels which have 
 been used for assays exist in the state of oxide, for no part 
 of their mass is free it is found even in the bottom ; 
 besides it is known that the lead carbonate precipitated 
 from lead acetate made from litharge contains silver, and a 
 notable quantity of that metal is found even in the lead 
 sulphate prepared by means of alum from the acetate 
 (unless the sulphate is repeatedly washed with water). 
 
 It has been remarked that the centres of cupels which 
 have been used for assays are richer in silver than the parts 
 near the circumference, and that under the button there is 
 a spot of bright yellow, which appears to be a compound 
 of silver. But the most important cause of loss in an assay 
 is the property which the alloys of silver and lead possess 
 of introducing themselves into the pores of the cupel. The 
 quantity thus lost is in proportion to the coarseness of the 
 cupel. For the same quantity of silver, the loss which 
 takes place in an assay varies according to the nature of 
 the alloy, and the circumstances under which the assay is 
 made ; so that it is not possible to form accurate tables of 
 correction. This loss is much augmented with the quantity 
 of lead employed, but without its being proportionate ; so 
 that when scorification is had recourse to, it is advantageous 
 to continue the operation for some length of time, in order 
 that the metallic button may be reduced to the smallest 
 suitable volume. 
 
 In the assay of rich alloys, the proportion to the total 
 amount of silver is very small, but notable ; and it has 
 been calculated for the alloys of copper employed in the 
 arts at ^ J^th ; but in the assay of poor ores, such as 
 galena and other minerals treated in the large way, the 
 loss is very great, for it is usually as high as -5^-0 th. 
 
 By extracting the lead from cupels used in this class 
 of assay, the metal furnished contains from about g- - oVo~o tn 
 
CUPELLATION. 625 
 
 to -s-innnnrti 1 f silver. The following experiment will 
 give an idea of the influence of the proportion of lead on 
 the loss of silver : 100 grains of commercial litharge were 
 fused with 10 grains of black flux, and gave 27 grains of 
 lead, and a slag ; this was pulverised and reduced in the 
 same crucible with 15 grains of black flux, and a second 
 button was produced weighing 45 grains. These two 
 buttons, being cupelled separately, gave, the first -0035 
 and the second -001 only of silver. Three new quantities 
 of 100 grains of the same litharge were fused ; the first 
 with -J a part of starch, the second with 2^, and the third 
 with 10 of the same reducing agent. The resulting 
 buttons of lead weighed respectively 5, 28, and 79 grains. 
 These buttons were cupelled, and furnished "0035, '0035, 
 and -003 respectively. From these experiments it will be 
 seen that when the litharge is not reduced completely, 
 there remains a notable proportion of silver in the scoriae ; 
 but, nevertheless, in order to extract the largest possible 
 quantity, the whole must not be reduced. Indeed, but a 
 twentieth part only need be reduced, because more precious 
 metal is lost in the cupellation of a large quantity of lead 
 than remains in the portion not reduced. The loss of 
 silver in large cupellations is less than that which takes 
 place in an assay, because in the large way the litharge, 
 or the greater part of it, is run off; whilst in an assay the 
 cupel totally absorbs it, so that the latter presents, rela- 
 tively to the same mass of lead, a very much smaller 
 surface in the large than in the small way ; now it can be 
 readily seen that the quantity of silver lost by absorption 
 into the pores of the cupel must be proportioned to its 
 surface, all things being equal. 
 
 It has been ascertained by experiment that a cupel 
 absorbs about its own weight of litharge ; so that from 
 this fact a cupel of the proper size may be chosen, when 
 the weight of lead to be cupelled is ascertained. It is 
 always better to have the cupel about ^ or \ as heavy 
 again as the lead to be cupelled. 
 
 The various metals found in an alloy, which can be 
 submitted to cupellation, scorify in proportion to their 
 
 s s 
 
626 THE ASSAY OF SILVER. 
 
 oxidisability. Those most oxidisable scorify with the 
 greatest rapidity, and vice versa ; so that those which 
 have the greatest affinity for oxygen accumulate in the 
 first portions of litharge formed, which, by that means 
 becoming less fusible, sometimes lose the property of 
 penetrating the cupel ; hence the reason why cupellations 
 always present more difficulties at the commencement of 
 the operation than towards the end, when the litharge 
 formed is nearly pure lead oxide, and can contain only 
 copper oxide. 
 
 The appearance of the cupel used in an assay will 
 give indications of the metals the alloy contained. Pure 
 lead colours the cupel straw-yellow, verging on lemon- 
 yellow. Bismuth, straw- yellow passing into orange-yellow. 
 Copper gives a grey, dirty red, or brown, according to its 
 proportion. Iron gives black scorise, which form at the 
 commencement of the operation, and are generally found 
 at the circumference of the -cupel. Tin gives a grey slag. 
 Zinc leaves a yellowish ring on the cupel, producing a 
 very luminous flame, and occasioning losses by carrying 
 silver in its vapour, and by projecting it from the cupel 
 in its ebullition. Antimony and lead sulphate in excess 
 give litharge-yellow scoriae, which crack the cupel ; but, 
 when not produced in too great a proportion, are gradually 
 absorbed by the litharge. If the lead alloy submitted to 
 cupellation is found to produce this effect, a fresh portion 
 must be mixed with its own weight of lead and scorified : 
 the button so obtained can now be cupelled. 
 
 Amalgamation. There are a certain number of argen- 
 tiferous matters which can be assayed by amalgamation, 
 as they are treated in the large way by that method. 
 Amongst these are native silver, chlorides, sulphides, and 
 arsenio-sulphides, which contain neither lead nor copper. 
 
 But this process is seldom had recourse to, because 
 it is long, troublesome, and less exact than those just 
 described. 
 
ASSAY OF THE ALLOYS OF SILVER AND COPPER. 627 
 
 Substances of the Second Class. 
 
 Native silver. 
 
 Alloys of copper and silver. 
 
 Alloys of other metals and silver 
 
 (artificial). 
 Silver antimonide. 
 
 Silver telluride. 
 
 Auriferous silver telluride (see 
 
 gold). 
 
 Silver hydrargyride (amalgam), 
 Silver auride (see gold). 
 
 Silver arsenide. 
 
 The following method of separating silver from galena 
 is given in the ' Chemical News,' vol ii. p. 239. 
 
 ' Galena consists, as is well known, of lead sulphide, 
 mixed with a variable proportion of silver sulphide, and 
 both these substances fuse together, or melt at a bright 
 red heat. Now, it so happens that, when silver sulphide 
 is fused with lead chloride, what is called a double de- 
 composition takes place ; that is to say, silver chloride 
 and lead sulphide are formed. Consequently, if we 
 fuse together a quantity of argentiferous galena and lead 
 chloride, we shall remove the whole of the silver from the 
 galena, and replace it by lead sulphide. This, then, is the 
 process : mix together the galena and lead chloride in the 
 proportion of 100 parts of galena, 1 part of lead chloride, 
 and 10 parts of sodium chloride or common salt ; or, if the 
 galena be very argentiferous, add a larger amount of lead 
 chloride. The whole is then fused together, when the 
 silver chloride and common salt rise to the surface, and 
 may be skimmed off, and the desilverised galena falls and 
 may be run out from the bottom. The mixture of silver 
 chloride and salt may then be decomposed by lime and 
 charcoal, or in any other manner, so as to reduce the silver 
 and a portion of the surplus lead chloride, by which a metal- 
 lic mass will result, suitable for the operation of the cupel.' 
 
 General Remarks on the Assay of the Alloys of Silver 
 and Copper. The assay of these alloys is nearly always 
 accomplished (at least in England) by cupellation. This 
 assay is most important, as it is by the results obtained in 
 the manner hereafter described that the price or value of 
 all kinds of silver bullion is decided. 
 
 This class of cupellation is effected without difficulty, 
 because the copper oxide forms so slowly, that the litharge 
 
 s s 2 
 
628 
 
 THE ASSAY OF SILVER. 
 
 is always enabled to pass it into the body of the cupel. 
 After having weighed the lead and placed it in the cupel, 
 as soon as it is perfectly fused place in it the alloy to be 
 assayed, wrapped either in blotting-paper or thin leaf-lead. 
 It is essential, in this class of assay, to employ a sufficient 
 quantity of lead to carry away all the copper. We may 
 always be sure of succeeding, whatever the alloy may be, 
 by employing the maximum proportion of lead, that is to 
 say, the quantity necessary to pass pure copper ; but as 
 the loss which the silver undergoes increases with the 
 length of the operation and with the mass of the oxidised 
 matters, it is indispensable to reduce this loss as much as 
 possible by reducing the proportion of lead to that which 
 is strictly necessary. Long experience has proved that 
 silver opposes the oxidation of copper by its affinity, so 
 that it is necessary to add a larger amount of lead in pro- 
 portion to the quantity of silver present. 
 
 M. D'Arcet has obtained the following results by 
 accurate experiments : 
 
 Standard of 
 silver 
 
 Quantity of 
 copper alloyed 
 
 Quantity of lead 
 necessary 
 
 Relation of lead 
 to copper 
 
 . 1000 
 
 
 
 lo*hs 
 
 
 950 
 
 50 
 
 3 
 
 eotoi 
 
 900 
 
 100 
 
 7 
 
 701 
 
 800 
 
 200 
 
 10 
 
 50 1 
 
 700 
 
 300 
 
 12 
 
 40 1 
 
 600 
 
 400 
 
 14 35 1 
 
 500 
 
 500 
 
 16 to 17 321 
 
 400 
 
 600 
 
 16 17 
 
 27 1 
 
 300 
 
 700 
 
 16 17 
 
 23 1 
 
 200 
 
 800 
 
 16 17 
 
 20 1 
 
 100 
 
 900 
 
 16 17 
 
 18 1 
 
 pure copper 
 
 1000 
 
 16 17 
 
 16 1 
 
 It is remarkable that below the standard of 500, the 
 same proportion of lead must be employed, whatever that 
 of copper. This fact is repeatedly verified by experiment. 
 Whenever fine silver is fused in a cupel, it is always neces- 
 sary to add lead, in order to cause the button to unite 
 and form well. If less than fV ns f ^ eac ^ be employed, 
 the button will be badly formed ; the litharge cannot 
 separate but by the action of a very strong heat, and a 
 
ASSAY OF THE ALLOYS OF SILVER AND COPPER. 
 
 629 
 
 considerable loss of silver ensues. If, on the contrary, 
 Y 3 ths of lead is exceeded, the cupellation goes on well, but 
 the loss is greater on account of the duration of the process. 
 These proportions also ought to vary with the temperature. 
 M. Chaudet has found that, to cupel an alloy containing 
 T 9 ^^ths of silver, 5 parts of lead are required in the middle 
 of the muffle, 10 in the front, and only 3 at the back. 
 
 The proportion of copper carried off by litharge varies 
 not only with the temperature, but even for the same 
 temperature in relation to the amount of copper and lead 
 the alloy contains. By cupelling 100 parts of copper with 
 different proportions of lead in the same furnace, M. 
 Karsten obtained the following results : 
 
 Lead added 
 
 Copper remaining after 
 cupellation 
 
 Quantity of lead consumed 
 in carrying off 1 of copper 
 
 100 
 
 78-75 
 
 3- 
 
 200 
 
 70-12 
 
 7-1 
 
 300 
 
 60-12 
 
 7-7 
 
 400 
 
 49-40 
 
 7-9 
 
 500 
 
 38-75 
 
 8-1 
 
 600 
 
 26-25 
 
 8-15 
 
 700 
 
 19-75 
 
 8-00 
 
 800 
 
 8-75 
 
 8-70 
 
 900 
 
 5-62 
 
 9-50 
 
 1000 
 
 1-25 
 
 10-10 
 
 1050 
 
 o-oo 
 
 10-50 
 
 From which we see that the lead carried away from -j^th 
 to yVth of its weight of copper. Much less lead can be 
 employed in a cupellation by making the alloy maintain 
 its richness of copper throughout the operation. This 
 can be accomplished by adding to the alloy in the cupel 
 small doses of lead, in proportion as that first added dis- 
 appears by oxidation. If, for example, an alloy composed 
 of 4 parts of copper and one of silver be fused with 10 of 
 lead, by adding successive small doses of the latter, as 
 already pointed out, but 7 parts will be consumed, although 
 in the regular way from 16 to 17 would be employed. 
 
 The proportion of copper oxide contained in the litharge 
 increases each instant, and goes on incessantly increasing 
 when an alloy of copper and lead is cupelled which con- 
 tains an excess of copper. According to M. Karsten, this 
 
(330 
 
 THE ASSAY OF SILVER. 
 
 proportion is always about 13 per cent, at the commence- 
 ment, and 36, or more than a third, at the end of the 
 operation. 
 
 In the assay of the coined alloys of copper and silver, 
 the loss of silver may even amount to five thousandths ; 
 but the loss is variable, and is proportionately greater as 
 the standard of the alloy is lower. 
 
 The following Table contains the results of many 
 experiments made on this subject : 
 
 
 
 Loss, or the quantity of fine 
 
 Exact standard 
 
 Standard found by cupellation 
 
 metal to be added to the stan- 
 dard as obtained by cupellation 
 
 1000 
 
 998-97 
 
 1-03 
 
 975 
 
 973-24 
 
 1-76 
 
 950 
 
 947-50 
 
 2-50 
 
 925 
 
 921-75 
 
 3-25 
 
 900 
 
 896-00 
 
 4-00 
 
 875 
 
 870-93 
 
 4-07 
 
 850 
 
 845-85 
 
 4-13 
 
 825 
 
 820-78 
 
 4-22 
 
 800 
 
 795-70 
 
 4-30 
 
 775 
 
 770-59 
 
 4-41 
 
 750 
 
 745-38 
 
 4-52 
 
 725 
 
 720-36 
 
 4-64 
 
 700 
 
 695-25 
 
 4-75 
 
 675 
 
 670-27 
 
 4-73 
 
 650 
 
 645-29 
 
 4-71 
 
 625 
 
 620-30 
 
 4-70 
 
 600 
 
 595-32 
 
 4-68 
 
 575 
 
 570-32 
 
 4-68 
 
 550 
 
 545-32 
 
 4-68 
 
 . 525 
 
 520-32 
 
 4-68 
 
 500 
 
 495-32 
 
 4-68 
 
 475 
 
 470-50 
 
 4-50 
 
 450 
 
 445-69 
 
 4-31 
 
 425 
 
 420-87 
 
 4-13 
 
 400 
 
 396-05 
 
 3-95 
 
 875 
 
 371-39 
 
 3-61 
 
 350 
 
 346-73 
 
 3-27 
 
 325 
 
 322-06 
 
 2-94 
 
 300 
 
 297-40 
 
 2-60 
 
 275 
 
 272-42 
 
 2-58 
 
 250 
 
 247-44 
 
 2-56 
 
 225 
 
 222-45 
 
 2-55 
 
 200 
 
 197-47 
 
 2-55 
 
 175 
 
 173-88 
 
 2-12 
 
 150 
 
 148-30 
 
 1-70 
 
 125 
 
 123-71 
 
 1-29 
 
 100' 
 
 99-12 
 
 0-88 
 
 75 
 
 74-34 
 
 0-66 
 
 50 ! 
 
 49-56 
 
 0-44 
 
 25 
 
 24-78 
 
 0-22 
 
ASSAY PKOPER OF SILVEE BULLION. 631 
 
 These numbers, however, are not constant, and vary with 
 the circumstances under which the assays are made : two 
 assays made from the same ingot, by the same assayer, 
 may differ as much as four or five thousandths. Tillet 
 has remarked that the cupels can retain double as much 
 silver as is lost ; which proves, as has already been men- 
 tioned, that the silver obtained by cupellation is not per- 
 fectly pure, but may retain as much as 1 per cent, of lead. 
 
 Special Instructions for the Assay of the Alloys of Silver 
 and Copper. 
 
 As before stated, peculiar weights are employed in the 
 assay of silver bullion ; and the silver assay pound, with 
 its divisions, will be found described at pages 34-35. 
 
 In the ' General Eemarks on the Assay of the Alloys of 
 Silver and Copper,' it will be seen that the alloy must be 
 cupelled with a quantity of lead, varying with the amount 
 of copper present in the alloy. Standard silver cupels very 
 well with five times its weight of lead ; but when the 
 approximate composition of the alloy is not known, it 
 must be estimated by a preliminary assay. 
 
 Assay for Approximative Composition of Alloy. Weigh 
 off 50 grains of pure or test lead ; place them in a cupel 
 previously made red-hot ; when the ]ead is fused, and its 
 surface covered with oxide, place in it by means of the 
 light tongs (a, fig. 28, page 68) 2 grains of the alloy under 
 assay, wrapped in a small piece of thin paper. Allow the 
 cupellation to go on according to the instructions, and with 
 all the precautions already given, and when complete, 
 weigh the resulting button, and, according to its weight, 
 add lead in the actual assay in the quantity that is suffi- 
 cient, as exhibited in the Table at page 628. 
 
 Assay Proper of Silver Bullion. In this assay the ope- 
 rator requires silver known to be standard, and pure lead. 
 With the possession of the above substances the assay is 
 thus proceeded with : Place the 12 grains weight(=l Ib.) 
 in the scale pan, and exactly counterbalance it with stan- 
 dard silver. This is to serve as a check. Remove the 
 weight, and in its place add so much of the alloy to be 
 
632 THE ASSAY OF SILVER. 
 
 assayed that the balance is again equal. In one cupel, that 
 destined to receive the check sample, place 60 grains of lead ; 
 and in another cupel place such a number of grains of lead 
 as may be found necessary by the preliminary assay. When 
 the lead in both cupels is fused, add the silver alloy, and 
 cupel with the necessary precautions. When the buttons in 
 the cupels are cold, seize them with the pliers, and if neces- 
 sary cleanse them with a hard brush, and place one in each 
 balance-pan. If they exactly balance each other, the alloy 
 operated on is standard silver ; if, however, it weighs less 
 than the button produced from the check sample by the 
 weight equivalent to 2 pennyweights, then it is 2 penny- 
 weights worse than standard : on the other hand, if it be 
 heavier by the same weight, it is 2 pennyweights better than 
 standard. Silver is also reported as so much fine : thus 
 standard silver may be reported as 11 ounces 2 penny- 
 weights fine, and so on. In case extreme accuracy be re- 
 quired, correction must be made according to the standard 
 as shown by the Table at page 630. The standard silver 
 in England is T V^- fine. 
 
 Assay of Alloys of Copper and Silver. In the treat- 
 ment on the large scale of copper ores containing silver, 
 the contained silver is found alloyed with the copper, and 
 it often falls under the assayer's province to estimate the 
 quantity of precious metal. An assay of this kind is most 
 conveniently accomplished by scorification before cupel- 
 lation, thus : Prepare four scorifiers ; weigh into each of 
 them 50 grains of the alloy, 50 grains of fused borax, and 
 600 grains of lead, and proceed as already described under 
 the head ' Assay of Ores of the First Class by Scorifica- 
 tion.' When the four buttons of lead are obtained, place 
 them together in another scorifier, and submit to the 
 furnace until the contents of the scorifier are completely 
 covered with oxide ; pour as usual, and cupel the re- 
 sulting mass of lead. 
 
 Alloys of Platinum and Silver. -If any substance con- 
 taining platinum as well as silver were assayed as already 
 described, the button resulting from the cupellation would, 
 in addition to the silver, contain the whole of the platinum. 
 
SEPARATING SILVER FROM THE BASER METALS. 633 
 
 In such a case the button so obtained must be thus 
 treated : 
 
 If the alloy contain much platinum, it must be fused 
 with twice its weight of silver ; then treated with hot nitric 
 acid ; evaporate the solution nearly to dryness ; add water 
 and hydrochloric acid, until no further precipitation of 
 silver as a white curdy precipitate (silver chloride) takes 
 place. The silver chloride may be collected either on a 
 filter or by decantation. The solution containing the 
 platinum is treated with excess of sal-ammoniac solution 
 until no further precipitation takes place ; the solution 
 evaporated to dryness. When cold, dilute alcohol is added ; 
 and the insoluble yellow matter (platinum ammonio- 
 chloride) collected on a filter, washed with alcohol, dried, 
 and ignited. The ignited residue is metallic platinum, 
 which is weighed. The loss of weight which the alloy 
 from cupel has sustained represents the amount of silver 
 previously alloyed with it. 
 
 Alloy of Platinum, Silver, and Copper. Treat such an 
 alloy as above ; and the liquid, filtered from the platinum 
 ammonio-chloride, will contain the copper. Acidulate it 
 with hydrochloric acid, add metallic zinc, and proceed as 
 directed under the head ' Wet Copper Assay.' 
 
 Native Silver, Rough Silver left on Sieve during Pulveri- 
 sation of Silver Ores of First Class, and Native Alloys of 
 Silver as Antimonides, c. are treated by scorification 
 and cupellation in precisely the same manner as just de- 
 scribed for alloys of copper and silver. 
 
 Dr. W. Dyce proposed, in ' Tilloch's Philosophical 
 Magazine ' for 1805, the following process for separating 
 gold and silver from the baser metals : 
 
 ' Hitherto the process has always been, as far as I have 
 understood it, attended with considerable difficulty in the 
 execution ; but, by that which I am about to describe, it 
 is done with exact certainty. It was discovered and com- 
 municated to me by a gentleman in the neighbourhood. 
 The process consists in mixing not less than two parts of 
 powdered manganese with the impure or compound metal, 
 which should be previously flattened or spread out so as 
 
634 THE ASSAY OF SILVEE. 
 
 to expose as large a surface as possible, and broken or cut 
 into small pieces for the convenience of putting the whole 
 into a crucible, which is then to be kept in a sufficient 
 heat for a short time. On removing the whole from the 
 fire, and allowing it to cool, the mixture is found to be 
 converted into a brownish powder, which powder or oxide 
 is then to be mixed with an equal proportion of powdered 
 glass, and then submitted in a crucible to a sufficient heat, 
 so as to fuse the whole, when the perfect metals are 
 found at the bottom in a state of extreme purity a cir- 
 cumstance of no small importance to the artist and the 
 chemist, the latter of whom will find no difficulty in sepa- 
 rating the one from the other, with so little trouble com- 
 pared with the usual processes that I have no doubt it 
 will always be practised in preference to the cupel.' 
 
 Assay of Silver Bullion by the Wet Method. From 
 what has been stated under the head of ' Cupellatiori,' it 
 will be observed that there are many sources of error ; 
 such as volatilisation of the precious metal, its oxidation 
 in the presence of excess of lead oxide and atmospheric 
 oxygen, and, lastly, its absorption into the body of the 
 cupel either as oxide or metal, or in both states. These 
 losses, as before stated, vary with the temperature, the 
 amount of lead employed, and the texture of the cupel ; 
 and, as may be seen from the table of corrections as 
 drawn up by D'Arcet, give a very erroneous assay, unless 
 the addition necessary for each standard be made. 
 
 Considerable attention was called to this matter in 
 France some years since, and a Special Commission was 
 appointed to examine the subject thoroughly, and, if 
 possible, to devise some means of assay which might be 
 both easy and accurate. The result of this examination 
 was the invention of a process of assay at once elegant 
 and trustworthy : and as a full account of this method has 
 not, to the author's knowledge, been translated and pub- 
 lished in this country,* he has prepared the present from 
 M. Gay-Lussac's Eeport, which formed a part of a com- 
 
 * Some portion of this report has been published in Dr. Ure's ' Dictionary 
 of Arts, Mines, and Manufactures.' 
 
ASSAY OF SILVER BULLION IN THE WET WAY. 635 
 
 munication from M. Thiers to Earl Granville, and which 
 appeared in the original language in the year 1837, in a 
 Eeport on the Eoyal Mint. 
 
 The process of assay about to be described consists in 
 estimating the fineness of silver bullion by the quantity 
 of a standard solution of common salt necessary to fully 
 and exactly precipitate the silver contained in a known 
 weight of alloy. This process is based on the following 
 principles : 
 
 The alloy, previously dissolved in nitric acid, is mixed 
 with a standard solution of common salt, which precipi- 
 tates the silver as chloride, a compound perfectly insoluble 
 in w r ater, and even in acids. 
 
 The quantity of silver chloride precipitated is esti- 
 mated not by its weight, which would be less exact and 
 occupy too much time, but by the weight or volume of 
 the standard solution of common salt necessary to exactly 
 precipitate the silver previously dissolved in nitric acid. 
 
 The term of complete precipitation of the silver can 
 be readily recognised by the cessation of all cloudiness 
 when the salt solution is gradually poured into that 
 of the nitrate of silver. One milligramme of that metal 
 is readily detected in 150 grammes of liquid ; and even 
 a half or a quarter of a milligramme may be detected, 
 if the liquid be perfectly bright before the addition of 
 the salt solution. 
 
 By violent agitation during a minute or two, the liquid, 
 rendered milky by the precipitation of silver chloride, 
 becomes sufficiently bright after a few moments' repose to 
 allow of the effect of the addition of half a milligramme of 
 silver to be perceptible. Filtration of the liquid is more 
 efficacious than agitation ; but the latter, which is much 
 more rapid, generally suffices. The presence of copper ,. 
 lead, or any other metal, with the exception of mercury 
 (the presence of the latter metal requires a slight modifi- 
 cation of the process, which will be hereafter pointed out) r 
 in the silver solution, has no sensible influence on the 
 quantity of salt required for precipitation : in other words, 
 the same quantity of silver, pure or alloyed, requires for 
 
636 THE ASSAY OF SILVER. 
 
 its precipitation a constant quantity of the standard salt 
 solution. 
 
 Supposing that 1 gramme of pure silver be the quan- 
 tity operated on, the solution of salt required to exactly 
 precipitate the whole of the silver ought to be of such 
 strength that, if it be measured by weight, it shall weigh 
 exactly 100 grammes, or if by volume 100 cubic centi- 
 metres. This quantity of salt solution is divided into 
 1000 parts, called thousandths. 
 
 The standard of an alloy of silver is generally the 
 number of thousandths of solution of salt necessary to 
 precipitate the silver contained in a gramme of the alloy. 
 Measurement of the Solution of Common Salt. The 
 solution of common salt will hereafter be termed the 
 normal solution of common salt. It can be measured by 
 weight or volume. The measure by weight gives greater 
 FIG ill precision, and it has the special advan- 
 
 tage of being independent of temperature ; 
 but it requires too much time in nume- 
 rous assays. The measure by volume 
 gives a sufficient exactitude, and requires 
 much less time than the measure by 
 weight ; it is, indeed, liable to the in- 
 fluence of temperature, but tables for 
 correction will be appended. 
 
 Measure of the Normal Solution of 
 Salt by Weight This solution should be 
 so made that 100 grammes will exactly 
 precipitate 1 gramme of pure silver 
 dissolved in nitric acid. In order to 
 point out the method of taking the 
 weight it must be supposed to have 
 been previously prepared. After the 
 process of taking the weight is described, 
 the mode of preparing the solution will 
 be given. 
 
 The solution is weighed in a burette 
 (fig. Ill) whose capacity is from 115 to 120 grammes of 
 the solution, and divided into grammes. These divisions are 
 
MEASUKEMENT OF THE SOLUTION OF COMMON SALT. 637 
 
 for the purpose of approximate! vely estimating the weight 
 of solution, so as to shorten the operation of weighing. The 
 burette is represented as closed by a cork, B, in order to 
 prevent evaporation of the solution when the instrument is 
 not in use. It is also easy to remedy the inconvenience of 
 evaporation, by rinsing the burette with a small quantity 
 of the fresh solution. On pouring the solution from the 
 orifice, (9, of the burette, each division will furnish from 
 8 to 10 drops ; and consequently the weight of a drop is 
 about a decigramme. The burette is filled with solution 
 to the division o ; it is then tared in a balance capable 
 of turning with a centigramme. The burette is then re- 
 moved, and its place supplied with a weight equivalent to 
 the amount of solution required 100 grammes, for in- 
 stance. The solution is then gradually poured from the 
 burette into a bottle appointed for its reception, until the 
 equilibrium is nearly established. It is not easy to attain 
 the point exactly, as no smaller quantity than a drop can 
 be poured from the burette. This, however, is a matter 
 of indifference ; it suffices to know the exact weight of 
 the solution poured out : suppose it to be 99 gr. 85 c. 
 The mode of more nearly approximating the 
 required weight of 100 grammes will now be 
 pointed out. 
 
 It must be remarked that it is not the 
 amount of water contained in the 100 grammes 
 that is of consequence, but only the quantity 
 of salt found in solution ; this should exactly 
 represent 1000 thousandths of pure silver. If 
 now 100 grammes of the normal solution be 
 mixed with 900 grammes of water, it is evi- 
 dent that 1 gramme of this new solution is 
 equivalent to a decigramme of the first, and 
 consequently it will be easy to obtain 100 
 grammes of the normal solution, or rather the 
 1000 thousandths of salt it ought to contain ; 
 it will now be sufficient to add to the 99 
 grammes already poured from the burette, 1 gramme 
 of the new solution. It can be weighed, like the normal 
 
 FIG. 112. 
 
038 . THE ASSAY OF SILVER. 
 
 solution, to a drop nearly, in the burette (fig. 112), which 
 is of such a diameter that each small division represents 
 a decigramme of liquid, and consequently a centigramme 
 of the normal solution ; but it is more readily measured 
 by volume, preparing it in the manner to be hereafter 
 pointed out. To avoid all confusion, a solution to be 
 termed a decime solution of common salt is one containing 
 the same quantity of salt as the normal solution, in a 
 weight or volume ten times greater. 
 
 A decime solution of silver is a solution of silver equi- 
 valent to the latter, both mutually suffering complete 
 decomposition. 
 
 Preparation of the Decime Solution of Common Salt. 
 One hundred grammes of the normal solution of common 
 
 FIG. 113. FIG. 114. 
 
 salt are weighed in a flask (fig. 113) containing a kilo- 
 gramme of pure water, when filled up to the mark a b, 
 or 1000 cubic centimetres ; this quantity is made up with 
 pure water, taking care to agitate the whole well, to 
 render the mixtuise, homogeneous. A cubic centimetre of 
 this solution represents 1 thousandth of silver. This 
 quantity is readily obtained by means of a pipette (fig. 
 114), gauged so that when filled up with water to the 
 mark c d, it shall allow 1 gramme, or 1 cubic centimetre, 
 to run freely, the small quantity of liquid remaining in 
 the pipette not forming part of the gramme. In pouring 
 the liquid by drops, a little more or a little less than 
 
PREPARATION OP THE DECIME SOLUTION OF SALT. 639 
 
 twenty may be counted, according to the size of the 
 orifice, o. This number will not vary more than one 
 drop. Half a cubic centimetre will consequently be re- 
 presented by 10 drops, and a quarter by 5. The precision 
 arrived at by this method of measurement suffices, since 
 the possible error on the cubic centimetre will be but one- 
 twentieth of that quantity, or one-twentieth of a thou- 
 sandth ; if, however, many measures be required, then 
 compensation must be made. 
 
 The decime solution of common salt requisite for assays 
 must be kept in a bottle (fig, 114) closed by a cork, tra- 
 versed by the pipette, firmly fixed in a FlG< 115< 
 hole bored for that purpose. To mea- 
 sure a thousandth with the pipette, 
 the bottle is held with one hand, and 
 the pipette with the other (fig. 115). 
 The pipette is taken from the solution 
 after its upper orifice has been closed 
 by the forefinger ; the lower orifice 
 is then inclined against the edge of 
 the flask to remove the liquid, which 
 without this precaution would remain 
 there : the mark c d is then raised to 
 the level of the eye, and by a suitable 
 pressure of the forefinger on the upper 
 orifice, which may be obtained by giving the pipette a 
 slight alternating circular movement between the fingers, 
 the solution is allowed to run out gradually. The instant 
 the concave surface of the liquid is at the level c d, the 
 pipette is firmly closed by pressure of the forefinger on its 
 orifice, which is held above the bottle into which the solu- 
 tion is to be poured, and the forefinger removed so that it 
 can be emptied. It is here necessary to remark that in 
 order to regulate the slow and regular runnings of the 
 liquid from the pipette, by the pressure of the forefinger, 
 the latter ought to be neither too moist nor too dry : if too 
 dry, it will not perfectly close the orifice, even by strong 
 pressure ; if too moist, it prevents the entrance of air, and 
 the liquid will not run, or if it do, it will be irregularly. 
 
640 THE ASSAY OF SILVER. 
 
 This observation should not be lost sight of in the use of 
 the large burettes mentioned hereafter. 
 
 Preparation of the Decime Solution of Silver.*- The 
 decline solution of silver is prepared by dissolving 1 
 gramme of pure silver in nitric acid, in a flask holding 1 
 litre (see fig. 113), and then diluting the solution with 
 distilled water, so that, cooled at the ordinary temperature 
 
 FIG. 116. 
 
 of the air, it shall occupy exactly the volume of one litre. 
 It is measured in precisely the same manner as the decime 
 salt solution. 
 
 Weighing the Normal Solution of Common Salt. To 
 execute this operation with rapidity, a balance similar 
 to that represented at fig. 116 is employed. The arms 
 are divided as in the assay balance described at p. 28 ; 
 each of the arms, c B, is furnished with a rider, c, of 
 such a weight (about 5 decigrammes) that, moved from the 
 right or the left of the centre, c, of each arm, it indicates 
 
PKEPAKATIOX OF THE NORMAL SALT SOLUTION. 641 
 
 two decigrammes. The space traversed by the rider is 
 divided into twenty equal parts, representing an equal 
 number of centigrammes. 
 
 We will take for example the weighing of 100 grammes 
 of normal solution of common salt, which is that most fre- 
 quently made in the estimation of the standard of all 
 varieties of argentiferous matter. 
 
 There are two weights, one, P, equal to the tare of the 
 burette when full of solution to the mark o, the other, P', 
 equals 100 grammes. The burette is filled with solution, 
 and placed on the right-hand pan of the balance, on which 
 it is kept in position by the collar d 0, and through which 
 it is passed before placing it on the pan. The tare, P, of 
 the burette is- supposed to be on the opposite side. If the 
 equilibrium be not perfect, it is effected by the rider on the 
 left ; the burette is then removed, and 100 grammes of the 
 solution (either more or less to one or two decigrammes) 
 poured out. The burette is then again placed in the 
 balance, with the 100-gramme weight P', the upper part 
 of which is slightly concave to receive the bottom of the 
 burette, in order to prevent it sliding off. The equilibrium 
 is again established by the aid of the rider on the right, 
 If, for instance, it is found necessary to remove the rider, 
 which represents 15 centigrammes, 15 divisions towards B, 
 the weight of the solution poured out of the burette will 
 be equal to 100 gr. - 0-15 gr. = 99'85 gr. If, on the other 
 hand, it is necessary to move the rider six divisions towards 
 c, the weight of the solution will be 100 gr. + 0-06 
 gr.=100-06 gr. 
 
 The above method of weighing the salt solution ap- 
 pears to be the most convenient that can be employed, 
 although it is not very expeditious. Other methods of 
 weighing and measuring will be given in an appendix to 
 this article. 
 
 Preparation of the Normal Solution of Common Salt 
 when measured by weight. After having pointed out the 
 method of weighing the normal solution of salt, and of 
 taking very small quantities, its preparation will be de- 
 scribed. 
 
 T T 
 
042 THE ASSAY OP SILVER. 
 
 Supposing the salt as well as the water to be employed 
 are pure, the two substances have only to be taken in the 
 following proportions : 0-5427 kilogramme of salt and 
 99*4573 kilogrammes of water, to form 100 kilogrammes 
 of solution, of which 100 grammes will exactly precipitate 
 1 gramme of silver. But instead of pure salt, which is 
 difficult to procure, and which besides rapidly alters by 
 the absorption of atmospheric moisture, it is preferable to 
 employ a concentrated solution of commercial salt, which 
 can be prepared in large quantities, and kept for use as 
 needed. The quantity of salt it contains can be ascer- 
 tained " s by evaporating a portion to dryness, and by a few 
 experiments it is easy to estimate in what proportion 
 it shall be mixed with water to produce a solution, 100 
 FIG 117 grammes of which shall exactly precipitate 
 
 1 gramme of silver. 
 
 Suppose, for example, that the salt 
 solution contains 250 grammes of salt per 
 kilogramme, and that it is necessary to 
 prepare 100 kilogrammes of the normal 
 solution. Now, since for the preparation 
 of this quantity 0-5427 kilogramme of pure 
 salt is required, we have the following pro- 
 portion : 
 
 0-25 : 1 :: 0-5427 : ^=2-1708 kilogs. 
 
 To this last weight enough water is added 
 to make up 100 kilogrammes, that is to say, 
 97-8292 kilogrammes, which quantity can 
 be readily measured by means of a flask 
 containing 5 or 6 kilogrammes previously 
 gauged. 
 
 The mixture must be well agitated by 
 means of the agitator (fig. 117), which is 
 made of an osier twig, split into four 
 branches, to the extremities of which is attached a small 
 square piece of silk. This'substance is employed to avoid 
 the separation of filaments which would ensue from the 
 use of any other material. This agitator can be intro- 
 
PREPARATION OP THE NORMAL SALT SOLUTION. G43 
 
 duced into very small openings, and is exceedingly ser- 
 viceable in agitating large masses of liquid. 
 
 When well mixed, the solution must be assayed. To 
 effect this, dissolve 1 gramme of silver in nitric acid, sp. 
 gr. 1-290, in a stoppered bottle (fig. 118) Fm. us. 
 holding about 200 grammes of water, tare 
 the burette (fig. 1] 1) filled with the solution, 
 and pour rather more than less into the 
 bottle ; in proportion as the salt employed is 
 impure, more than 100 grammes will be re- 
 quired to precipitate 1 gramme of silver. The 
 mixture is at first milky, but, by vigorously 
 shaking the bottle, having its stopper firmly 
 fixed, for about a minute, and then allowing it 
 to remain at rest for a short time, the liquid will become 
 perfectly bright : two drops of the solution must then be 
 poured into it from the burette : if a cloudiness is pro- 
 duced, it is agitated again to brighten it, and two drops 
 more added. This must be continued until the last two 
 drops added give no precipitate. The operation is then 
 terminated, and nothing remains to state but the result. 
 
 Supposing the total weight of solution poured from the 
 burette is 101-88 grammes, the last two drops must not 
 be reckoned, because they produce no effect ; the two pre- 
 ceding drops were necessary, but in part only ; that is 
 to say, the number of drops to be deducted is less than 
 four, and more than two, or rather it is the mean term, 
 three. Or the weight of a drop can be known exactly 
 by taking that of a dozen: suppose it is equal to 0*082 
 gramme, three times that number must be deducted, 
 or 0*255 grammes from 101-880 grammes: there will 
 remain 101-625 grammes, representing the quantity of 
 normal solution necessary to precipitate 1 gramme of 
 silver. 
 
 The solution is thus found to be too weak ; to bring it 
 to its proper standard it is necessary to remove 1-625 
 gramme of water from the 101-625 grammes of solution, 
 or, what is the same thing, to add to the normal solution 
 a certain quantity of the concentrated solution of common 
 
 T T 2 
 
644 THE ASSAY OF SILVER. 
 
 salt, which quantity may be found by the following pro- 
 portion : 
 
 100 : -1-625 :: 2-1708 kilogrs. of silver solution : ^ = 0*0353. 
 
 After the addition of this quantity of salt to the normal 
 solution, a fresh assay is made, proceeding in precisely the 
 same manner as before ; taking care, however, to pour 
 from the burette a weight of solution slightly under 100 
 grammes, or 1000 decigrammes ; for instance, 998*4 
 decigrammes, because it is not possible, in pouring the 
 solution by drops, to arrive at the exact weight, 1000 
 decigrammes. To ascertain the true standard in the most 
 exact manner possible, a decime solution must be prepared 
 by weighing 100 grammes of the normal solution, and 
 diluting it with pure water, so that it shall occupy one 
 litre : a cubic centimetre of this solution will represent a 
 decigramme of the normal solution. This decime solution 
 will not be rigorously exact, since the normal solution has 
 not been truly standardised ; but it is easily perceived 
 that the error thus committed is very small, and that it 
 may be neglected. Nevertheless, as soon as the normal 
 solution is perfectly standardised, it is better to prepare 
 another decime solution. 
 
 A decime solution may be immediately obtained by 
 dissolving 0*5427 gramme of pure sea-salt in such a quan- 
 tity of water that the whole will occupy one litre ; yet the 
 first process is preferable. 
 
 With the decime solution the assay may be thus con- 
 tinued, remembering that the pipette described at fig. 114 
 is a cubic centimetre containing 20 drops ; that the half, 
 therefore, is represented by 10 drops, and the fourth by 5. 
 
 To the 998-4 decigrammes of normal solution already 
 added, pour one pipette and 12 drops of the decime 
 solution, which will exactly complete the weight of 1000 
 decigrammes of normal solution. The mixture is agitated 
 to brighten it, and one-thousandth of common salt or one 
 pipette of the decime solution added. If this causes a 
 cloudiness, it is agitated and a second thousandth added. 
 This last should produce no opalescence. The weight 
 
PREPARATION OF THE NORMAL SALT SOLUTION. 645 
 
 of normal solution necessary to exactly precipitate one 
 gramme of silver will be between 1000 and 1001 deci- 
 grammes ; that is to say, the mean will be equal to 1000-5. 
 The standard of the normal solution is then too weak by 
 half a thousandth ; to correct this a quantity of concen- 
 trated salt solution must be added equal to half a thou- 
 sandth of that already added (2-1708 + 0-0353 = 2-2061 
 kilogrammes) ; that is to say, 1-1 gramme. 
 
 A new assay is then made for verification. 
 
 When the standard of a solution is very nearly arrived 
 at, it is well to employ filtration to detect the slightest 
 opalescence, at least when sufficient time is not .allowed for 
 the liquid to become perfectly bright. The surest method, 
 when the standard is nearly attained, is to place some of 
 the liquid in two test-glasses, and pour into one a few 
 drops of the decime solution of common salt, and into the 
 other a corresponding number of drops of the decime 
 solution of silver nitrate. It may then be determined on 
 which side the opalescence is manifested, and the assay of 
 the normal solution may be continued after the mixture of 
 the liquid in the two glasses, since the two quantities of 
 the decime solutions of common salt and silver nitrate 
 mutually decompose each other, and do not interfere with 
 the assay. Once the standard of the normal solution is 
 definitely fixed, the sum of the quantities of the concen- 
 trated solution of common salt which have been employed, 
 as well as those of the water, must be noted, and in the 
 preparation of a new normal solution the proportions 
 found as above would only have to be mixed to obtain at 
 once a solution having very nearly its true standard. 
 
 In determining the standard of the normal solution, 
 supposing that it were always too weak, it would be 
 necessary to add to the solution a certain quantity of 
 common salt ; but if the true amount had been exceeded, 
 and it had been found too strong, the solution would 
 have to be precipitated with the decime solution of silver ; 
 and knowing the number of cubic centimetres or thou- 
 sandths of silver which had been necessary to ; precipitate 
 the excess of common salt, it could be estimated what 
 
646 THE ASSAY OF SILVER. 
 
 amount of water must be added to reduce the normal 
 solution to standard. For instance, if 2 thousandths of 
 the decime solution of silver had been consumed, 2 thou- 
 sandths of its weight of water would have to be added 
 to the total amount of solution ; that is to say, 0-2 kilo- 
 gramme or 200 grammes. 
 
 Preservation of the Normal Solution of Common Salt. 
 The most suitable vessel for containing the normal solution 
 of common salt is one of glass, because that cannot affect 
 the standard. Large glass bottles, termed carboys, are 
 found in commerce. These bottles contain from 50 to 60 
 litres, and are very applicable for this purpose. Fig. 119 
 represents one of these bottles fixed in a stand formed of a 
 sieve hoop. It is graduated into litres or kilogrammes of 
 water, and a paper scale fixed on its side shows at any 
 time the quantity of contained liquid. It is closed by an 
 hydraulic valve, made of sheet iron, but the bell or cover 
 is of glass. The detail of this valve is shown at fig. 120. 
 FIG. 119. The a ^ r can on ty en ter the bottle by 
 
 the narrow tube T 7 , and cannot pass 
 out by it ; consequently evaporation 
 is not to be feared. The neck of 
 the valve should be about a deci- 
 metre deep, into which mercury 
 should be poured, but only to about 
 one-third of its height. 
 
 The solution is drawn from the 
 bottle by the syphon S. This is 
 furnished with a stopcock ; but this 
 syphon being brittle, at least when 
 not of metal, is not convenient in 
 use, since it is incorporated with the- 
 bell of the valve : it is, therefore, 
 preferable to pierce the bottom of 
 the bottle (fig. 121), and fix a metal tube (T) by means of 
 a plate moulded on the bottom and cemented to it. This 
 tube is raised a little above the bottom of the bottle, and 
 covered by a small cup, the object of which is to protect 
 it from any of the mercury which might fall into it. It is 
 
APPLICATION OF GAY-LUSSAC's PKOCESS. 
 
 647 
 
 terminated at its other extremity by a very narrow tube, 
 so that the flow of the solution may not be too rapid- 
 
 FIG. 120. 
 
 FIG. 121. 
 
 Fifl. 122. 
 
 Hereafter a metal reservoir will be described which has 
 all the advantages of a glass vessel without its inconve- 
 niences. 
 
 Application of the Process described in the Estimation 
 of the Standard of a Silver Alloy. The alloy is supposed 
 to be that made into coin, the mean standard 
 of which is fixed at 900 thousandths, but 
 which may vary from 897 to 903 thousandths 
 without ceasing to be legal (French standard 
 for coin). One gramme is dissolved in the 
 bottle (fig. 118) by about 10 grammes of 
 nitric acid, sp. gr. 1-290. This quantity of 
 nitric acid can be readily taken by means of 
 the pipette P (fig. 122), which contains 7'7 
 grammes of water to the mark a b. The 
 solution may be accelerated by placing the 
 bottle in a small pan of hot water, the bottom 
 of which must be covered with a piece of 
 cloth, so as to prevent contact of the glass and metal. 
 The solution finished, and the flask slightly cooled, the 
 
648 THE ASSAY OF SILVER. 
 
 nitrous vapour must be removed 4 by a blower (see fig. 123), 
 the nozzle of which is formed of a piece of bent glass tube, 
 connected by a cork with a copper socket 
 D, having a screw inside. This operation 
 ought to be effected, as well as the solution 
 of the alloy in nitric acid, under a chimney 
 with a strong current of air, to carry off the 
 nitrous vapour. 
 
 The burette (fig. 99), being filled with 
 the normal solution of common salt, and 
 tared, about 90 grammes are poured into the 
 solution of the alloy ; say 89 -85 grammes. 
 After agitating the liquor, a cubic centimetre 
 of the decime solution of common salt is added 
 representing one-thousandth of silver. If a cloudiness 
 be observed, agitate again, and add a second thousandth 
 of common salt, and so on, until the last thousandth 
 gives no precipitate. Suppose it to be the fourth : 
 that must not be counted, because it has produced no 
 effect ; and only a half of the third must be taken, 
 because only a portion of that was necessary. The 
 standard of the alloy would be consequently equal to 
 
 If it be desirable to approach still nearer to the true 
 standard of the alloy, half-thousandths must be added until 
 the last half-thousandth gives no precipitate ; and in order 
 to avoid all confusion, it is better to write with chalk on 
 a blackboard the thousandths of common salt, preceding 
 them by the plus sign +, and on the other side the thou- 
 sandths of the silver nitrate, preceding them by the sign 
 minus. 
 
 In the above example, after the addition of the 4 
 thousandths of common salt, the last of which has pro- 
 duced no cloudiness, 1^ thousandth of nitrate of silver is 
 added, which destroys 1^ thousandth of common salt, and 
 brightens the liquid. If another half-thousandth of nitrate 
 of silver praduce no precipitate, it is not taken into account, 
 and is struck off from the table. From whence it is con- 
 cluded that the quantity of nitrate of silver necessary to 
 
APPLICATION OF GAY-LUSSAC'S PROCESS. 649 
 
 destroy the excess of common salt is more than 1 and less 
 than li ; that is to say, nearly the J of a thousandth, and 
 is equal to 1J. Thus the number of thousandths of salt 
 really used is 41-25 = 275. The standard of the alloy, 
 therefore, is 898-50 + 275 = 901-25. 
 
 Another example, everything else remaining as above : 
 Suppose the first thousandth of salt did not precipitate. 
 This is a proof that too much normal solution of common 
 salt has been employed, and that there is an excess of salt 
 in the liquid. Add one- thousandth of silver, and agitate : 
 things are now as at first, but it is nevertheless known 
 that it is with nitrate of silver the process must be con- 
 tinued. One-thousandth has been added, which produced 
 a precipitate ; the second does not. The standard of the 
 .alloy is consequently 898-5 0-5 = 898. To approach still 
 nearer to the real standard, destroy the last 2 thousandths 
 of silver by 2 thousandths of common salt, and add half a 
 thousandth of silver a cloudiness is produced, as already 
 known ; but another half-thousandth does not precipitate. 
 The standard of the alloy is therefore 898-50-0-25 = 
 898-25. 
 
 This process, on which it would be useless to enlarge 
 further at present, because many other parts of the pro- 
 cess to be presently described apply to it, is general, and 
 gives exactly the standard of an alloy when it is known 
 approximately, which can always be ascertained by a 
 previous rough assay. 
 
 Correction of the Standard of the Normal Solution of 
 Salt when the Temperature Varies. It has been assumed 
 that, in the estimation of the standard of the normal 
 solution of salt, the temperature has remained constant. 
 Assays made under these circumstances need no correc- 
 tion ; but if the temperature changes, the same measure 
 of solution will not contain the same amount of salt. Sup- 
 posing the solution of salt has been standardised at 15. 
 If, at the time an experiment is made, the temperature 
 is 18, for instance, the solution will be found too weak, 
 since it has expanded, and the pipette holds less than its 
 proper weight. If, on the other hand, the temperature 
 
650 THE ASSAY OF SILVEE. 
 
 falls to 12, .the solution becomes concentrated, and is 
 found too strong. It is therefore necessary to estimate 
 the correction to be made for any variation of temperature 
 that may occur. 
 
 To this end the temperature of a solution of common 
 salt has been gradually raised from 0... .5. ...10... .15. ...20.... 
 25.. ..30 degrees, and three pipettefuls of the solution 
 exactly weighed at each of the above temperatures. One- 
 third of the total weight gives the mean weight of the 
 contents of a pipette. The corresponding weights of a 
 pipetteful of solution are then entered, and form the second 
 column of the following table, called ' Table of Correction 
 for the Variations of Temperature in the Normal Solution 
 of Salt/ By this table correction may be made for any 
 temperature between and 30, when the solution of 
 salt has been standardised within the same limits. Sup- 
 pose, for example, the solution had been standardised at 
 15, and that at the time it was used its temperature was 
 18. On referring to the second column of the table, it 
 will be seen that the weight o a measure of solution at 
 15 is 100-099 gr. ; and at 18, 100-065 gr. ; the difference 
 0-034 gr. is the quantity of solution taken too little, and 
 consequently it must be added to the normal measure, so 
 that it may be equal to one thousand thousandths. If the 
 temperature of the solution had fallen to 10, the differ- 
 ence of weight between a measure at 10 and a measure at 
 15 will be 0-019 gr., which must, on the contrary, be de- 
 ducted from the measure, as it has been taken in excess. 
 These differences of weight of a measure of solution at 15, 
 and that of a measure for any other temperature, form the 
 column 15 in the table, where they are expressed in thou- 
 sandths. They are written on the same horizontal line as the 
 temperatures to which each corresponds, with the sign + 
 when they are to be added, and the sign when to be 
 subtracted. The columns 5, 10, 20, 25, 30 have been 
 calculated in the same manner, to meet cases in which the 
 normal solution had been graduated at each of the above- 
 named temperatures. Thus, to calculate the column 10, 
 take the number 100-118 from the column of weights as 
 
CORRECTION OF THE STANDARD OF SOLUTION OF SALT. 651 
 
 a point of departure, and find the difference for all the 
 other numbers in the same column. 
 
 An application of this Table will be given hereafter. 
 
 TABLE OF CORRECTIONS FOR VARIATIONS IN TEMPERATURE OF THE NORMAL 
 
 SALT SOLUTION. 
 
 Temperature 
 
 Weight 
 
 5 
 
 10 
 
 15 
 
 20 
 
 25 
 
 30 
 
 Degrees 
 
 Grammes 
 
 Mill. 
 
 Mill. 
 
 Mill. 
 
 Mill. 
 
 Mill. 
 
 Mill. 
 
 4 100-109 
 
 o-o 
 
 -o-i 
 
 + 0-1 
 
 + 07 
 
 + 1-7 
 
 + 2-7 
 
 5 100-113 
 
 o-o 
 
 -0-1 
 
 + 0-1 
 
 + 0-7 
 
 + 1-7 
 
 + 2-8 
 
 6 100-115 
 
 o-o 
 
 o-o 
 
 + 0-2 
 
 40-8 
 
 hl-7 
 
 + 2-8 
 
 7 100-118 
 
 + 0-1 
 
 o-o 
 
 + 0-2 
 
 + 0-8 
 
 + 1-7 +2-8 
 
 8 100-120 
 
 + 0-1 
 
 o-o 
 
 + 0-2 
 
 + 0-8 
 
 + 1-8 
 
 + 2-8 
 
 9 100-120 
 
 + 0-1 
 
 o-o 
 
 + 0-2 
 
 + 0-8 
 
 + 1-8 
 
 + 2-8 
 
 10 100-118 
 
 + 0-1 
 
 o-o 
 
 + 0-2 
 
 + 0-8 
 
 4-1-7 
 
 + 2-8 
 
 11 100-116 
 
 o-o 
 
 o-o 
 
 + 0-2 
 
 + 0-8 
 
 + 1-7 
 
 + 2-8 
 
 12 
 
 100-114 
 
 o-o 
 
 o-o 
 
 + 0-2 
 
 + 0-8 
 
 + 1-7 
 
 + 2-8 
 
 13 
 
 100-110 
 
 o-o 
 
 -o-i 
 
 + 0-1 
 
 + 0-7 
 
 + 1-7 
 
 + 2-7 
 
 14 
 
 100-106 
 
 -0-1 
 
 -0-1 +0-1 
 
 + 0-7 +1-6 
 
 + 2-7 
 
 15 
 
 100-099 
 
 -0-1 
 
 -0-2 0-0 
 
 + 0-6 +1-6 
 
 + 2-6 
 
 16 
 
 100-090 
 
 -0-2 
 
 -0-3 -0-1 
 
 + 0-5 + 1-5 
 
 + 2-5 
 
 17 
 
 100-078 
 
 -0-4 
 
 -0-4 
 
 -0-2 
 
 + 0-4 
 
 + 1-3 
 
 + 2-4 
 
 18 
 
 100-065 
 
 -0-5 
 
 -0-5 -0-3 
 
 + 0-3 
 
 + 1-2 
 
 + 2-3 
 
 19 
 
 100-053 
 
 -o-o 
 
 -0-7 
 
 -0-5 
 
 + 0-1 
 
 + 1-1 
 
 + 2-2 
 
 20 
 
 100-039 
 
 -0-7 
 
 -0-8 
 
 -0-6 
 
 0-0 
 
 + 1-0 
 
 4-2-0 
 
 21 
 
 100-021 
 
 -0-9 
 
 -1-0 
 
 -0-8 
 
 -0-2 
 
 + 0-8 
 
 + 1-9 
 
 22 
 
 100-001 
 
 -M 
 
 -1-2 
 
 -1-0 
 
 -0-4 
 
 + 0-6 
 
 4-1-7 
 
 23 
 
 99-983 
 
 -1-3 
 
 -1-4 
 
 -1-2 -0-6 
 
 + 0-4 
 
 + 1-5 
 
 24 99-964 
 
 -1-5 
 
 -1-5 
 
 -1-4 
 
 -0-8 
 
 + 0-2 
 
 + 1-3 
 
 25 99-944 
 
 -1-7 
 
 -1-7 
 
 -1-6 1 -1-0 
 
 o-o 
 
 + 1-1 
 
 26 
 
 99-924 
 
 -1-9 
 
 -1-9 
 
 -1-8 1 -1-2 -0-2 
 
 + 0-9 
 
 27 
 
 99-902 
 
 -2-1 
 
 -2-2 
 
 -2-0 -1-4 
 
 -0-4 
 
 + 0-7 
 
 28 
 
 99-879 
 
 -2-3 
 
 -2-4 
 
 -2-2 
 
 -1-6 
 
 -0-7 
 
 + 0-4 
 
 29 
 
 99-858 
 
 -2-6 
 
 -2-6 
 
 -2-4 
 
 -1-8 
 
 -0-9 
 
 -t-0-2 
 
 30 
 
 99-836 
 
 -2-8 
 
 -2-8 
 
 -2-6 
 
 -2-0 
 
 -1-1 
 
 0-0 
 
 TABLE FOR THE ASSAY, BY THE WET METHOD, OF AN ALLOY 
 CONTAINING ANY PROPORTIONS WHATEVER OF SILVER, BY 
 THE EMPLOYMENT OF A CONSTANT MEASURE OF THE NORMAL 
 SOLUTION OF COMMON SALT. 
 
 The process by which the normal solution of salt is 
 measured by weight is applicable to the assay of every 
 kind of alloy, since it suffices to take a weight of the solu- 
 tion corresponding to the presumed standard of the silver, 
 and complete the assay by means of the decime solu- 
 tion ; the process by volume, however, has not the same 
 
652 THE ASSAY OF SILVER. 
 
 advantage, because the volume of normal solution cannot 
 be varied in the same manner as the weight. This incon- 
 venience, however, is of no very great consequence, for, 
 by keeping the volume of normal solution constant, it 
 suffices to vary the weight of the alloy, taking in each 
 particular case a weight which contains approximative] y 
 one gramme of pure silver. Suppose the alloy has a 
 standard of about 900 thousandths, we have the following 
 proportion : 
 
 900 thousandths : 1000 of alloy :: 1000 thousandths: 
 
 x = 1111-1. 
 
 If that weight be now taken to ascertain the standard 
 of the alloy, it may be found, for instance, that to the mea- 
 sure of 1000 thousandths of salt it is yet necessary to add 
 4 thousandths of salt to precipitate the whole of the silver ; 
 that is to say, that 11.11-1 of alloy really contain 1004 of 
 silver. From this result the real standard of the alloy may 
 be found to be 903-6, by the following equation : 
 
 1111-1 : 1004 : : 1000 : x = 903-6. 
 
 But such calculations, however simple, should be 
 .avoided where numerous daily assays are made, not only 
 on account of the time consumed, but still more from the 
 errors to which such operations are necessarily exposed. 
 Fortunately, all these inconveniences may be avoided by 
 the use of tables, which entirely free the assay er from 
 calculation. 
 
 Wishing in weighing the alloy to avoid fractions of 
 thousandths, and only making use of tenths and half-tenths 
 of thousandths, the weight of alloy increases, starting from 
 a gramme, from 5 to 5 thousandths, and the correspond- 
 ing standard for each of these weights has been sought, all 
 containing one gramme of pure silver. Thus the weight 
 1020 of alloy, in which there are 1000 of silver and 20 of 
 copper, corresponds to the standard 980-39, obtained by 
 the proportion 
 
 1020 : 1 000 : : 1000 : x = 980-39. 
 
EXPLANATION OF THE FOLLOWING TABLES. 653 
 
 On this principle are formed the first and second 
 columns of the table marked Salt. The first contains the 
 weight of each alloy, and the second its corresponding 
 standard. The following columns, 1, 2, 3, to 10, give the 
 standard of the alloy, when, instead of the 1000 milli- 
 grammes of silver it was supposed to contain, it really 
 contained 1, 2, 3, &c. more, and consequently 1, 2, 3, (fee- 
 milligrammes of copper less. 
 
 Another table, constructed in the same manner as the 
 preceding, and marked Nitrate of Silver, gives the standard 
 of the alloy when, under the weight given in the first 
 column, it contains 1, 2, 3, <fec. milligrammes less silver, 
 and as much more copper. Thus, for example, an alloy of 
 the weight of 1020 (1000 silver and 20 copper) has for its 
 standard 980-4 in both tables. If it always contains in the 
 same weight 4 more silver and consequently 4 less copper, 
 its standard would be 984*3, and would be found in the 
 4 Salt ' table at the intersection of the column 4, and the 
 horizontal line 1020. If, on the contrary, it contains 4 
 less of silver and 4 more of copper, its standard will be 
 976-5, and will be found in the 'Nitrate of Silver ' table, at 
 the intersection of the column 4, and the horizontal line 
 1020. 
 
654 
 
 THE ASSAY OF SILVER. 
 
 Tables for Estimating the Standard of any Silver 
 approximativcly containing 
 
 NITRATE OF 
 
 Weight of 
 
 
 
 
 
 
 Assay in 
 
 0. 
 
 i. 
 
 2. 
 
 3. 
 
 4. 
 
 Milligrs. 
 
 
 
 
 
 
 1000 
 
 1000-0 
 
 999-0 
 
 998-0 
 
 997-0 
 
 996-0 
 
 1005 
 
 995-0 
 
 994-0 
 
 993-0 
 
 992-0 
 
 991-0 
 
 1010 
 
 990-1 
 
 989-1 
 
 988-1 
 
 987-1 
 
 986-1 
 
 1015 
 
 985-2 
 
 984-2 
 
 983-2 
 
 982-3 
 
 981-3 
 
 1020 
 
 980-4 
 
 979-4 
 
 978-4 
 
 977-4 
 
 976-5 
 
 1025 
 
 975-6 
 
 974-6 
 
 973-7 
 
 972-7 
 
 971-7 
 
 1030 
 
 970-9 
 
 969-9 
 
 968-9 
 
 968-0 
 
 967-0 
 
 1035 
 
 966-2 
 
 965-2 
 
 964-2 
 
 963-3 
 
 962-3 
 
 1040 
 
 961-5 
 
 960-6 
 
 959-6 
 
 958-6 
 
 957-7 
 
 1045 
 
 956-9 
 
 956-0 
 
 955-0 
 
 954-1 
 
 953-1 
 
 1050 
 
 952-4 
 
 951-4 
 
 950-5 
 
 949-5 
 
 948-6 
 
 1055 
 
 947-9 
 
 946-9 
 
 946-0 
 
 945-0 
 
 944-1 
 
 1060 
 
 943-4 
 
 942-4 
 
 941-5 
 
 940-6 
 
 939-6 
 
 1065 
 
 939-0 
 
 938-0 
 
 937-1 
 
 936-1 
 
 935-2 
 
 1070 
 
 934-6 
 
 933-6 
 
 932-7 
 
 931-8 
 
 930-8 
 
 1075 
 
 930-2 
 
 929-3 
 
 928-4 
 
 927-4 
 
 926-5 
 
 1080 
 
 925-9 
 
 925-0 
 
 924-1 
 
 923-1 
 
 922-2 
 
 1085 
 
 921-7 
 
 920-7 
 
 919-8 
 
 918-9 
 
 918-0 
 
 1090 
 
 917-4 
 
 916-5 
 
 915-6 
 
 914-7 
 
 913-8 
 
 1095 
 
 913-2 
 
 912-3 
 
 911-4 
 
 910-5 
 
 909-6 
 
 1100 
 
 909-1 
 
 908-2 
 
 907-3 
 
 906-4 
 
 905-4 
 
 1105 
 
 905-0 
 
 904-1 
 
 903-2 
 
 902-3 
 
 901-4 
 
 1110 
 
 900-9 
 
 900-0 
 
 899-1 
 
 898-2 
 
 897-3 
 
 1115 
 
 896-9 
 
 896-0 
 
 895-1 
 
 894-2 
 
 893-3 
 
 1120 
 
 892-9 
 
 892-0 
 
 891-1 
 
 890-2 
 
 889-3 
 
 1125 
 
 888-9 
 
 888-0 
 
 887-1 
 
 886-2 
 
 885-3 
 
 1130 
 
 885-0 
 
 884-1 
 
 883-2 
 
 882-3 
 
 881-4 
 
 1135 
 
 881-1 
 
 880-2 
 
 879-3 
 
 878-4 
 
 877-5 
 
 1140 
 
 877-2 
 
 876-3 
 
 875-4 
 
 874-6 
 
 873-7 
 
 1145 
 
 873-4 
 
 872-5 
 
 871-6 
 
 870-7 
 
 869-9 
 
 1150 
 
 869-6 
 
 868-7 
 
 867-8 
 
 867-0 
 
 866-1 
 
 1155 
 
 865-8 
 
 864-9 
 
 864-1 
 
 863-2 
 
 862-3 
 
 1160 
 
 862-1 
 
 861-2 
 
 860-3 
 
 859-5 
 
 858-6 
 
 1165 
 
 858-4 
 
 857-5 
 
 856-6 
 
 855-8 
 
 854-9 
 
 1170 
 
 854-7 
 
 853-8 
 
 853-0 
 
 852-1 
 
 851-3 
 
 1175 
 
 851-1 
 
 850-2 
 
 849-4 
 
 848-5 
 
 847-7 
 
 1180 
 
 847-5 
 
 846-6 
 
 845-8 
 
 844-9 
 
 844-1 
 
 1185 
 
 843-9 
 
 843-0 
 
 842-2 
 
 841-3 
 
 840-5 
 
TABLE FOR THE WET ASSAY OF SILVER. 
 
 655 
 
 Alloy by employing an Amount of Alloy always 
 the same Amount of Silver. 
 
 SILVER. 
 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 
 995-0 
 
 994-0 
 
 993-0 
 
 992-0 
 
 991-0 
 
 990-0 
 
 
 990-0 
 
 989-0 
 
 988-1 
 
 987-1 
 
 986-1 
 
 985-1 
 
 
 985-1 
 
 984-2 
 
 983-2 
 
 982-2 
 
 981-2 
 
 980-2 
 
 
 980-3 
 
 979-3 
 
 978-3 
 
 977-3 
 
 976-4 
 
 975-4 
 
 
 975-5 
 
 974-5 
 
 973-5 
 
 972-5 
 
 971-6 
 
 970-6 
 
 
 970-7 
 
 969-8 
 
 968-8 
 
 967-8 
 
 966-8 
 
 965-8 
 
 
 966-0 
 
 965-0 
 
 964-1 
 
 963-1 
 
 962-1 
 
 961-2 
 
 
 961-3 
 
 960-4 
 
 959-4 | 958-4 
 
 957-5 
 
 956-5 
 
 
 956-7 
 
 955-8 
 
 954-8 
 
 953-8 
 
 952-9 
 
 951-9 
 
 
 952-1 
 
 951-2 
 
 950-2 
 
 949-3 
 
 948-3 
 
 947-4 
 
 
 947-6 
 
 946-7 
 
 945-7 
 
 944-8 
 
 943-8 
 
 942-9 
 
 
 943-1 
 
 942-2 
 
 941-2 
 
 940-3 
 
 939-3 
 
 938-4 
 
 
 938-7 
 
 937-7 
 
 936-8 
 
 935-8 
 
 934-9 
 
 934-0 
 
 
 934-3 
 
 933-3 
 
 932-4 
 
 931-4 
 
 930-5 
 
 929-6 
 
 
 929-9 
 
 929-0 
 
 928-0 
 
 927-1 
 
 926-2 
 
 925-2 
 
 
 925-6 
 
 924-7 
 
 923-7 
 
 922-8 
 
 921-9 
 
 920-9 
 
 
 921-3 
 
 920-4 
 
 919-4 
 
 918-5 
 
 917-6 
 
 916-7 
 
 
 917-0 
 
 916-1 
 
 915-2 
 
 914-3 
 
 913-4 
 
 912-4 
 
 
 912-8 
 
 911-9 
 
 911-0 
 
 910-1 
 
 909-2 
 
 908-3 
 
 
 908-7 
 
 907-8 
 
 906-8 
 
 905-9 
 
 905-0 
 
 904-1 
 
 
 904-5 
 
 903-6 
 
 902-7 
 
 901-8 
 
 900-9 
 
 900-0 
 
 
 900-4 
 
 899-5 
 
 898-6 
 
 897-7 
 
 896-8 
 
 895-9 
 
 
 896-4 
 
 895-5 
 
 894-6 
 
 893-7 
 
 892-8 
 
 891-9 
 
 
 892-4 
 
 891-5 
 
 890-6 
 
 889-7 
 
 888-8 
 
 887-9 
 
 
 888-4 
 
 887-5 
 
 886-6 
 
 885-7 
 
 884-8 
 
 883-9 
 
 
 884-4 
 
 883-6 
 
 882-7 
 
 881-8 
 
 880-9 
 
 880-0 
 
 
 880-5 
 
 879-6 
 
 878-8 
 
 877-9 
 
 877-0 
 
 876-1 
 
 
 876-7 
 
 875-8 
 
 874-9 
 
 874-0 
 
 873-1 
 
 872-3 
 
 
 872-8 
 
 871-9 
 
 871-0 
 
 870-2 
 
 869-3 
 
 868-4 
 
 
 869-0 
 
 868-1 
 
 867-2 
 
 866-4 
 
 865-5 
 
 864-6 
 
 
 865-2 
 
 864-3 
 
 863-5 
 
 862-6 
 
 861-7 
 
 860-9 
 
 
 861-5 
 
 860-6 
 
 859-7 
 
 858-9 
 
 858-0 
 
 857-1 
 
 
 857-8 
 
 856-9 
 
 856-0 
 
 855-2 
 
 854-3 
 
 853-4 
 
 
 854-1 
 
 853-2 
 
 852-4 
 
 851-5 
 
 850-6 
 
 849-8 
 
 
 850-4 
 
 849-6 
 
 848-7 
 
 847-9 
 
 847-0 
 
 846-1 
 
 
 846-8 
 
 846-0 
 
 845-1 
 
 844-3 
 
 843-4 
 
 842-5 
 
 
 843-2 
 
 842-4 
 
 841-5 
 
 840-7 
 
 839-8 
 
 839-0 
 
 
 839-7 
 
 838-8 
 
 838-0 
 
 837-1 
 
 836-3 
 
 835-4 
 
(156 
 
 THE ASSAY OF SILVER. 
 
 NITKATE OF 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 i. 
 
 2. 
 
 3. 
 
 4. 
 
 
 1190 
 
 840-3 
 
 849-5 
 
 838-7 
 
 837-8 
 
 837-0 
 
 
 1195 
 
 836-8 
 
 836-0 
 
 835-1 
 
 834-3 
 
 833-5 
 
 
 1200 
 
 833-3 
 
 832-5 
 
 831-7 
 
 830-8 
 
 830-0 
 
 
 1205 
 
 829-9 
 
 829-0 
 
 828-2 
 
 827-4 
 
 826-6 
 
 
 1210 
 
 826-4 
 
 825-6 
 
 824-8 
 
 824-0 
 
 823-1 
 
 
 1215 
 
 823-0 
 
 822-2 
 
 821-4 
 
 820-6 
 
 819-7 
 
 
 1220 
 
 819-7 
 
 818-8 
 
 818-0 
 
 817-2 
 
 816-4 
 
 
 1225 
 
 816-3 
 
 815-5 
 
 814-7 
 
 813-9 
 
 813-1 
 
 
 1230 
 
 813-0 
 
 812-2 
 
 811-4 
 
 810-6 
 
 809-8 
 
 
 1235 
 
 809-7 
 
 808-9 
 
 808-1 
 
 807-3 
 
 806-5 
 
 
 1240 
 
 806-5 
 
 805-6 
 
 804-8 
 
 804-0 
 
 803-2 
 
 
 1245 
 
 803-2 
 
 802-4 
 
 801-6 
 
 800-8 
 
 800-0 
 
 
 1250 
 
 800-0 
 
 799-2 
 
 798-4 
 
 797-6 
 
 796-8 
 
 
 1255 
 
 796-8 
 
 796-0 
 
 795-2 
 
 794-4 
 
 793-6 
 
 
 1260 
 
 793-6 
 
 792-9 
 
 792-1 
 
 791-3 
 
 790-5 
 
 
 1265 
 
 790-5 
 
 789-7 
 
 788-9 
 
 788-1 
 
 787-3 
 
 
 1270 
 
 787-4 
 
 786-6 
 
 785-8 
 
 785-0 
 
 784-2 
 
 
 1275 
 
 784-3 
 
 783-5 
 
 782-7 
 
 782-0 
 
 781-2 
 
 
 1280 
 
 781-2 
 
 780-5 
 
 779-7 
 
 778-9 
 
 778-1 
 
 
 1285 
 
 778-2 
 
 777-4 
 
 776-6 
 
 775-9 
 
 775-1 
 
 
 1290 
 
 775-2 
 
 774-4 
 
 773-6 
 
 772-9 
 
 772-1 
 
 
 1295 
 
 772-2 
 
 771-4 
 
 770-7 
 
 769-9 
 
 769-1 
 
 
 1300 
 
 769-2 
 
 768-5 
 
 767-7 
 
 766-9 
 
 766-1 
 
 
 1305 
 
 766-3 
 
 765-5 
 
 764-7 
 
 764-0 
 
 763-2 
 
 
 1310 
 
 763-4 
 
 762-6 
 
 761-8 
 
 761-1 
 
 760-3 
 
 
 1315 
 
 760-5 
 
 759-7 
 
 758-9 
 
 758-2 
 
 757-4 
 
 
 1320 
 
 757-6 
 
 756-8 
 
 756-1 
 
 755-3 
 
 754-5 
 
 
 1325 
 
 754-7 
 
 754-0 
 
 753-2 
 
 752-4 
 
 751-7 
 
 
 1330 
 
 751-9 
 
 751-1 
 
 750-4 
 
 749-6 
 
 748-9 
 
 
 1335 
 
 749-1 
 
 748-3 
 
 747-6 
 
 746-8 
 
 746-1 
 
 
 1340 
 
 746-3 
 
 745-5 
 
 744-8 
 
 744-0 
 
 743-3 
 
 
 1345 
 
 743-5 
 
 742-7 
 
 742-0 
 
 741-3 
 
 740-5 
 
 
 1350 
 
 740-7 
 
 740-0 
 
 739-3 
 
 738-5 
 
 737-8 
 
 
 1355 
 
 738-0 
 
 737-3 
 
 736-5 
 
 735-8 
 
 735-1 
 
 
 1360 
 
 735-3 
 
 734-6 
 
 733-8 
 
 733-1 
 
 732-4 
 
 
 1365 
 
 732-6 
 
 731-9 
 
 731-1 
 
 730-4 
 
 729-7 
 
 
 1370 
 
 729-9 
 
 729-2 
 
 728-5 
 
 727-7 
 
 727-0 
 
 
 1375 
 
 727-3 
 
 726-5 
 
 725-8 
 
 725-1 
 
 724-4 
 
 
 1380 
 
 724-6 
 
 723-9 
 
 723-2 
 
 722-5 
 
 721-7 
 
 
 1385 
 
 722-0 
 
 721-3 
 
 720-6 
 
 719-9 
 
 719-1 
 
 
 1390 
 
 719-4 
 
 718-7 
 
 718-0 
 
 717-3 
 
 716-5 
 
 
 1395 
 
 716-8 
 
 716-1 
 
 715-4 714-7 
 
 714-0 
 
 
 1400 
 
 714-3 
 
 713-6 
 
 712-9 712-1 
 
 711-4 
 
 
TABLE FOR THE WET ASSAY OF SILVER. 
 
 657 
 
 
 SILVER continued. 
 
 6. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 
 836-1 
 
 835-3 
 
 834-5 
 
 833-6 
 
 832-8 
 
 831-9 
 
 
 832-6 
 
 831-8 
 
 831-0 
 
 830-1 
 
 829-3 
 
 828-4 
 
 
 829-2 
 
 828-3 
 
 827-5 
 
 826-7 
 
 825-8 
 
 825-0 
 
 
 825-7 
 
 824-9 
 
 824-1 
 
 823-2 
 
 822-4 
 
 821-6 
 
 
 822-3 
 
 821-5 
 
 820-7 
 
 819-8 
 
 819-0 
 
 818-2 
 
 
 818-9 
 
 818-1 
 
 817-3 
 
 816-5 
 
 815-6 
 
 814-8 
 
 
 815-6 
 
 814-7 
 
 813-9 
 
 813-1 
 
 812-3 
 
 811-5 
 
 
 812-2 
 
 811-4 
 
 810-6 
 
 809-8 
 
 809-0 
 
 808-2 
 
 
 808-9 
 
 808-1 
 
 807-3 
 
 806-5 
 
 805-7 
 
 804-9 
 
 
 805-7 
 
 804-9 
 
 804-0 
 
 803-2 
 
 802-4 
 
 801-6 
 
 
 802-4 
 
 801-6 
 
 800-8 
 
 800-0 
 
 799-2 
 
 798-4 
 
 
 799-2 
 
 798-4 
 
 797-6 
 
 796-8 
 
 796-0 
 
 795-2 
 
 
 796-0 
 
 795-2 
 
 794-4 
 
 793-6 
 
 792-8 
 
 792-0 
 
 
 792-8 
 
 792-0 
 
 791-2 
 
 790-4 
 
 789-6 
 
 788-8 
 
 
 789-7 
 
 788-9 
 
 788-1 
 
 787-3 
 
 786-5 
 
 785-7 
 
 
 786-6 
 
 785-8 
 
 785-0 
 
 784-2 
 
 783-4 
 
 782-6 
 
 
 783-5 
 
 782-7 
 
 781-9 
 
 781-1 
 
 780-3 
 
 779-5 
 
 
 780-4 
 
 779-6 
 
 778-8 
 
 778-0 
 
 777-3 
 
 776-5 
 
 
 777-3 
 
 776-6 
 
 775-8 
 
 775-0 
 
 774-2 
 
 773-4 
 
 
 774-3 
 
 773-5 
 
 772-8 
 
 772-0 
 
 . 771-2 
 
 770-4 
 
 
 771-3 
 
 770-5 
 
 769-8 
 
 769-0 
 
 768-2 
 
 767-4 
 
 
 768-3 
 
 767-6 
 
 766-8 
 
 766-0 
 
 765-2 
 
 764-5 
 
 
 765-4 
 
 764-6 
 
 763-8 
 
 763-1 
 
 762-3 
 
 761-5 
 
 
 762-4 
 
 761-7 
 
 760-9 
 
 760-1 
 
 759-4 
 
 758-6 
 
 
 759-5 
 
 758-8 
 
 758-0 
 
 757-2 
 
 756-5 
 
 755-7 
 
 
 756-6 
 
 755-9 
 
 755-1 
 
 754-4 
 
 753-6 
 
 752-8 
 
 
 753-8 
 
 753-0 
 
 752-3 
 
 751-5 
 
 750-8 
 
 750-0 
 
 
 750-9 
 
 750-2 
 
 749-4 
 
 748-7 
 
 747-9 
 
 747-2 
 
 
 748-1 
 
 747-4 
 
 746-6 
 
 745-9 
 
 745-1 
 
 744-4 
 
 
 745-3 
 
 744-6 
 
 743-8 
 
 743-1 
 
 742-3 
 
 741-6 
 
 
 742-5 
 
 741-8 
 
 741-0 
 
 740-3 
 
 739-5 
 
 738-8 
 
 
 739-8 
 
 739-0 
 
 738-3 
 
 737-5 
 
 736-8 
 
 736-1 
 
 
 737-0 
 
 736-3 
 
 735-6 
 
 734-8 
 
 734-1 
 
 733-3 
 
 734-3 
 
 733-6 
 
 732-8 
 
 732-1 
 
 731-4 
 
 730-6 
 
 
 731-6 
 
 730-9 
 
 730-1 
 
 729-4 
 
 728-7 
 
 727-9 
 
 
 728-9 
 
 728-2 
 
 727-5 
 
 726-7 
 
 726-0 
 
 725-3 
 
 
 726-3 
 
 725-5 
 
 724-8 
 
 724-1 
 
 723-4 
 
 722-6 
 
 
 723-6 
 
 722-9 
 
 722-2 
 
 721-4 
 
 720-7 
 
 720-0 
 
 
 721-0 
 
 720-3 
 
 719-6 
 
 718-8 
 
 718-1 
 
 717-4 
 
 
 718-4 
 
 717-7 
 
 717-0 
 
 716-2 
 
 715-5 
 
 714-8 
 
 
 715-8 
 
 715-1 
 
 714-4 
 
 713-7 
 
 712-9 
 
 712-2 
 
 
 713-3 
 
 712-5 
 
 711-8 
 
 711-1 
 
 710-4 
 
 709-7 
 
 
 710-7 
 
 710-0 
 
 709-3 
 
 708-6 
 
 707-9 
 
 707-1 
 
 u u 
 
658 
 
 THE ASSAY OF SILVER. 
 
 NITRATE 01 
 
 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 i. 
 
 2. 
 
 3. 
 
 4. 
 
 1405 
 
 711-7 
 
 711-0 
 
 710-3 
 
 709-6 
 
 708-9 
 
 
 1410 
 
 709-2 
 
 708-5 
 
 707-8 
 
 707-1 
 
 706-4 
 
 
 1415 
 
 706-7 
 
 706-0 
 
 705-3 
 
 704-6 
 
 703-9 
 
 
 1420 
 
 704-2 
 
 703-5 
 
 702-8 
 
 702-1 
 
 701-4 
 
 
 1425 
 
 701-8 
 
 701-0 
 
 700-3 
 
 699-6 
 
 698-9 
 
 
 1430 
 
 699-3 
 
 698-6 
 
 697-9 
 
 697-2 
 
 696-5 
 
 
 1435 
 
 696-9 
 
 696-2 i 695-5 
 
 694-8 
 
 694-1 
 
 
 1440 
 
 694-4 
 
 693-7 
 
 693-1 
 
 692-4 
 
 691-7 
 
 
 1445 
 
 692-0 
 
 691-3 
 
 690-7 
 
 690-0 
 
 689-3 
 
 
 1450 
 
 689-7 
 
 689-0 
 
 688-3 
 
 687-6 
 
 686-9 
 
 
 1455 
 
 687-3 
 
 686-6 
 
 685-9 
 
 685-2 
 
 684-5 
 
 
 1460 
 
 684-9 
 
 684-2 
 
 683-6 
 
 682-9 
 
 682-2 
 
 
 1465 
 
 682-6 
 
 681-9 
 
 681-2 
 
 680-6 
 
 679-9 
 
 
 1470 
 
 680-3 
 
 679-6 
 
 678-9 
 
 678-2 
 
 677-5 
 
 
 1475 
 
 678-0 
 
 677-3 
 
 676-6 
 
 675-9 
 
 675-2 
 
 
 1480 
 
 675-7 
 
 675-0 
 
 674-3 
 
 673-6 
 
 673-0 
 
 
 1485 
 
 673-4 
 
 672-7 
 
 672-0 
 
 671-4 
 
 670-7 
 
 
 1490 
 
 671-1 
 
 670-5 
 
 669-8 
 
 669-1 
 
 668-5 
 
 
 1495 
 
 668-9 
 
 668-2 
 
 667-6 
 
 666-9 
 
 666-2 
 
 
 1500 
 
 666-7 
 
 666-0 
 
 665-3 
 
 664-7 
 
 664-0 
 
 
 1505 
 
 664-5 
 
 663-8 
 
 663-1 
 
 662-5 
 
 661-8 
 
 
 1510 
 
 662-3 
 
 661-6 
 
 660-9 
 
 660-3 
 
 659-6 
 
 
 1515 
 
 660-1 
 
 659-4 
 
 658-7 
 
 658-1 
 
 657-4 
 
 
 1520 
 
 657-9 
 
 657-2 
 
 656-6 
 
 655-9 
 
 655-3 
 
 
 1525 
 
 655-7 
 
 655-1 
 
 654-4 
 
 653-8 
 
 653-1 
 
 
 1530 
 
 653-6 
 
 652-9 
 
 652-3 
 
 651-6 
 
 651-0 
 
 
 1535 
 
 651-5 
 
 650-8 
 
 650-2 
 
 649-5 
 
 648-9 
 
 
 1540 
 
 649-4 
 
 648-7 
 
 648-0 
 
 647-4 
 
 646-7 
 
 
 1545 
 
 647-2 
 
 646-6 
 
 645-9 
 
 645-3 
 
 644-7 
 
 
 1550 
 
 645-2 
 
 644-5 
 
 643-9 
 
 643-2 
 
 642-6 
 
 
 1555 
 
 643-1 
 
 642-4 
 
 641-8 
 
 641-2 
 
 640-5 
 
 
 1560 
 
 641-0 
 
 640-4 
 
 639-7 
 
 639-1 
 
 638-5 
 
 
 1565 
 
 639-0 
 
 638-3 
 
 637-7 
 
 637-1 
 
 636-4 
 
 
 1570 
 
 636-9 
 
 636-3 
 
 635-7 
 
 635-0 
 
 634-4 
 
 
 1575 
 
 634-9 
 
 634-3 
 
 633-6 
 
 633-0 
 
 632-4 
 
 
 1580 
 
 632-9 
 
 632-3 
 
 631-6 
 
 631-0 
 
 630-4 
 
 
 1585 
 
 630-9 
 
 630-3 
 
 629-6 
 
 629-0 
 
 628-4 
 
 
 1590 
 
 628-9 
 
 623-3 
 
 627-7 
 
 627-0 
 
 626-4 
 
 
 1595 
 
 627-0 
 
 626-3 
 
 625-7 
 
 625-1 
 
 624-4 
 
 
 1600 
 
 625-0 
 
 624-4 
 
 623-7 
 
 623-1 
 
 622-5 
 
 
 1605 
 
 623-1 
 
 622-4 
 
 621-8 
 
 621-2 
 
 620-6 
 
 
 1610 
 
 621-1 
 
 620-5 
 
 619-9 
 
 619-2 
 
 618-6 
 
 
 1615 
 
 619-2 
 
 618-6 
 
 618-0 
 
 617-3 
 
 616-7 
 
 
TABLE FOE THE WET ASSAY OF SILVER. 
 
 
 SILVEK continued. 
 
 10. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 
 708-2 
 
 707-5 
 
 706-8 
 
 706-0 
 
 705-3 
 
 704-6 
 
 
 705-7 
 
 705-0 
 
 704-3 
 
 703-5 
 
 702-8 
 
 702-1 
 
 
 703-2 
 
 702-5 
 
 701-8 
 
 701-1 
 
 700-3 
 
 699-6 
 
 
 700-7 
 
 700-0 
 
 699-3 
 
 698-6 
 
 697-9 
 
 697-2 
 
 
 698-2 
 
 697-5 
 
 696-8 
 
 696-1 
 
 695-4 
 
 694-7. 
 
 
 695-8 
 
 695-1 
 
 694-4 
 
 693-7 
 
 693-0 
 
 692-3 
 
 
 693-4 
 
 692-7 
 
 692-0 
 
 691-3 
 
 690-6 
 
 689-9 
 
 
 691-0 
 
 690-3 
 
 689-6 
 
 688-9 
 
 688-2 
 
 687-5 
 
 
 688-6 
 
 687-9 
 
 687-2 
 
 686-5 
 
 685-8 
 
 685-1 
 
 
 686-2 
 
 685-5 
 
 684-8 
 
 684-1 
 
 683-4 
 
 682-8 
 
 
 683-8 
 
 683-2 
 
 682-5 
 
 681-8 
 
 681-1 
 
 680-4 
 
 
 681-5 
 
 680-8 
 
 680-1 
 
 679-4 
 
 678-8 
 
 678-1 
 
 
 679-2 
 
 678-5 
 
 677-8 
 
 677-1 
 
 676-4 
 
 675-8 
 
 
 676-9 
 
 676-2 
 
 675-5 
 
 674-8 
 
 674-1 
 
 673-5 
 
 
 674-6 
 
 673-9 
 
 673-2 
 
 672-5 
 
 671-9 
 
 671-2 
 
 
 672-3 
 
 671-6 
 
 670-9 
 
 670-3 
 
 669-6 
 
 668-9 
 
 
 670-0 
 
 669-4 
 
 668-7 
 
 668-0 
 
 667-3 
 
 666-7 
 
 
 667-8 
 
 667-1 
 
 666-4 
 
 665-8 
 
 665-1 
 
 664-4 
 
 
 665-5 
 
 664-9 
 
 064-2 
 
 663-5 
 
 662-9 
 
 662-2 
 
 
 663-3 
 
 662-7 
 
 662-0 
 
 661-3 
 
 660^7 
 
 660-0 
 
 
 661-1 
 
 660-5 
 
 659-8 
 
 659-1 
 
 658-5 
 
 657-8 
 
 
 658-9 
 
 658-3 
 
 657-6 
 
 656-9 
 
 656-3 
 
 655-6 
 
 
 656-8 
 
 656-1 
 
 655-4 
 
 654-8 
 
 654-1 
 
 653-5 
 
 
 654-6 
 
 653-9 
 
 653-3 
 
 652-6 
 
 652-0 
 
 651-3 
 
 
 652-5 
 
 651-8 
 
 651-1 
 
 650-5 
 
 649-8 
 
 649-2 
 
 
 650-3 
 
 649-7 
 
 649-0 
 
 648-4 
 
 647-7 
 
 647-1 
 
 
 648-2 
 
 647-6 
 
 646-9 
 
 646-2 
 
 645-6 
 
 644-9 
 
 
 646-1 
 
 645-4 
 
 644-8 
 
 644-2 
 
 643-5 
 
 642-9 
 
 
 644-0 
 
 643-4 
 
 642-7 
 
 642-1 
 
 641-4 
 
 640-8 
 
 
 641-9 
 
 641-3 
 
 640-6 
 
 640-0 
 
 639-3 
 
 638-7 
 
 
 639-9 
 
 639-2 
 
 638-6 
 
 637-9 
 
 637-3 
 
 636-7 
 
 
 637-8 
 
 637-2 
 
 636-5 
 
 635-9 
 
 635-3 
 
 634-6 
 
 
 635-8 
 
 635-1 
 
 634-5 
 
 633-9 
 
 633-2 
 
 632-6 
 
 
 633-8 
 
 633-1 
 
 632-5 
 
 631-8 
 
 631-2 
 
 630-6 
 
 
 631-7 
 
 631-1 
 
 630-5 
 
 629-8 
 
 629-2 
 
 628-6 
 
 
 629-7 
 
 629-1 
 
 628-5 
 
 627-8 
 
 627-2 
 
 626-6 
 
 
 627-8 
 
 627-1 
 
 626-5 
 
 625-9 
 
 625-2 
 
 624-6 
 
 
 625-8 
 
 625-2 
 
 624-5 
 
 623-9 
 
 623-3 
 
 622-6 
 
 
 623-8 
 
 623-2 
 
 622-6 
 
 621-9 
 
 621-3 
 
 620-7 
 
 
 621-9 
 
 621-2 
 
 620-6 
 
 620-0 
 
 619-4 
 
 618-7 
 
 
 619-9 
 
 619-3 
 
 618-7 
 
 618-1 
 
 617-4 
 
 616-1 
 
 
 618-0 
 
 617-4 
 
 616-8 
 
 616-1 
 
 615-5 
 
 614-9 
 
 
 616-1 
 
 615-5 
 
 614-9 
 
 614-2 
 
 613-6 
 
 613-0 
 
660 
 
 THE ASSAY OP SILVEK. 
 
 NITKATE OF 
 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 l. 
 
 2. 
 
 3. 
 
 4. 
 
 1620 
 
 617-3 
 
 616-7 
 
 616-0 
 
 615-4 
 
 614-8 
 
 
 1625 
 
 615-4 
 
 614-8 
 
 614-1 
 
 613-5 
 
 612-9 
 
 
 1630 
 
 613-5 
 
 612-9 
 
 612-3 
 
 611-7 
 
 611-0 
 
 
 1635 
 
 611-6 
 
 611-0 
 
 610-4 
 
 609-8 
 
 609-2 
 
 
 1640 
 
 609-8 
 
 609-1 
 
 608-5 
 
 607-9 
 
 607-3 
 
 
 1645 
 
 607-9 
 
 607-3 
 
 606-7 
 
 606-1 
 
 605-5 
 
 
 1650 
 
 606-1 
 
 605-4 
 
 604-8 
 
 604-2 
 
 603-6 
 
 
 1655 
 
 604-2 
 
 603-6 
 
 603-0 
 
 602-4 
 
 601-8 
 
 
 1660 
 
 602-4 
 
 601-8 
 
 601-2 
 
 600-6 
 
 600-0 
 
 
 1665 
 
 600-6 
 
 600-0 
 
 599-4 
 
 598-8 
 
 598-2 
 
 
 1670 
 
 598-8 
 
 598-2 
 
 597-6 
 
 597-0 
 
 596-4 
 
 
 1675 
 
 597-0 
 
 596-4 
 
 595-8 
 
 595-2 
 
 594-6 
 
 
 1680 
 
 595-2 
 
 594-6 
 
 594-0 
 
 593-4 
 
 592-9 
 
 
 1685 
 
 593-5 
 
 592-9 
 
 592-3 
 
 591-7 
 
 591-1 
 
 
 1690 
 
 591-7 
 
 591-1 
 
 590-5 
 
 589-9 
 
 589-3 
 
 
 1695 
 
 590-0 
 
 589-4 
 
 588-8 
 
 588-2 
 
 587-6 
 
 
 1700 
 
 588-2 
 
 587-6 
 
 587-1 
 
 586-5 
 
 585-9 
 
 
 1705 
 
 586-5 
 
 585-9 
 
 585-3 
 
 584-7 
 
 584-2 
 
 
 1710 
 
 584-8 
 
 584-2 
 
 583-6 
 
 583-0 
 
 582-5 
 
 
 1715 
 
 583-1 
 
 582-5 
 
 581-9 
 
 581-3 
 
 580-8 
 
 
 1720 
 
 581-4 
 
 580-8 
 
 580-2 
 
 579-6 
 
 579-1 
 
 
 1725 
 
 579-7 
 
 579-1 
 
 578-5 
 
 578-0 
 
 577-4 
 
 
 1730 
 
 578-0 
 
 577-5 
 
 576-9 
 
 576-3 
 
 575-7 
 
 
 1735 
 
 576-4 
 
 575-8 
 
 575-2 
 
 574-6 
 
 574-1 
 
 
 1740 
 
 574-7 
 
 574-1 
 
 573-6 
 
 573-0 
 
 572-4 
 
 
 1745 
 
 573-1 
 
 572-5 
 
 571-9 
 
 571-3 
 
 570-8 
 
 
 1750 
 
 571-4 
 
 570-9 
 
 570-3 
 
 569-7 
 
 569-1 
 
 
 1755 
 
 569-8 
 
 569-2 
 
 568-7 
 
 568-1 
 
 567-5 
 
 
 1760 
 
 568-2 
 
 567-6 
 
 567-0 
 
 566-5 
 
 565-9 
 
 
 1765 
 
 566-6 
 
 566-0 
 
 565-4 
 
 564-9 
 
 564-3 
 
 
 1770 
 
 565-0 
 
 564-4 
 
 563-8 
 
 563-3 
 
 562-7 
 
 
 1775 
 
 563-4 
 
 562-8 
 
 562-2 
 
 561-7 
 
 561-1 
 
 
 1780 
 
 561-8 
 
 561-2 
 
 560-7 
 
 560-1 
 
 559-5 
 
 
 1785 
 
 560-2 
 
 559-7 
 
 559-1 
 
 558-5 
 
 558-0 
 
 
 1790 
 
 558-7 
 
 558-1 
 
 557-5 
 
 557-0 
 
 556-4 
 
 
 1795 
 
 557-1 
 
 556-5 
 
 556-0 
 
 555-4 
 
 554-9 
 
 
 1800 
 
 555-6 
 
 555-0 
 
 554-4 
 
 553-9 
 
 553-3 
 
 
 1805 
 
 554-0 
 
 553-5 
 
 552-9 
 
 552-3 
 
 551-8 
 
 
 1810 
 
 552-5 
 
 551-9 
 
 551-4 
 
 550-8 
 
 550-3 
 
 
 1815 
 
 551-0 
 
 550-4 
 
 549-9 
 
 549-3 
 
 548-8 
 
 
 1820 
 
 549-4 
 
 548-9 
 
 548-3 
 
 547-8 
 
 547-2 
 
 
 1825 
 
 547-9 
 
 547-4 
 
 546-8 
 
 546-3 
 
 545-7 
 
 
 1830 
 
 546-4 
 
 545-9 
 
 545-4 
 
 544-8 
 
 544-3 
 
 
TABLE FOR THE WET ASSAY OF SILVER. 
 
 661 
 
 SILVER continued. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 614-2 
 
 613-6 
 
 613-0 
 
 612-3 
 
 611-7 
 
 611-1 
 
 612-3 
 
 611-7 
 
 611-1 
 
 610-5 
 
 609-8 
 
 609-2 
 
 610-4 
 
 609-8 
 
 609-2 
 
 608-6 
 
 608-0 
 
 607-4 
 
 608-6 
 
 607-9 
 
 607-3 
 
 606-7 
 
 606-1 
 
 605-5 
 
 606-7 
 
 606-1 
 
 605-5 
 
 604-9 
 
 604-3 
 
 603-7 
 
 604-9 
 
 604-3 
 
 603-6 
 
 603-0 
 
 602-4 
 
 601-8 
 
 603-0 
 
 602-4 
 
 601-8 
 
 601-2 
 
 600-6 
 
 600-0 
 
 601-2 
 
 600-6 
 
 600-0 
 
 599-4 
 
 598-8 
 
 598-2 
 
 599-4 
 
 598-8 
 
 598-2 
 
 597-6 
 
 597-0 
 
 596-4 
 
 597-6 
 
 597-0 
 
 596-4 
 
 595-8 
 
 595-2 
 
 594-6 
 
 595-8 
 
 595-2 
 
 594-6 
 
 594-0 
 
 593-4 
 
 592-8 
 
 594-0 
 
 593-4 
 
 592-8 
 
 592-2 
 
 591-6 
 
 591-0 
 
 592-3 
 
 591-7 
 
 591-1 
 
 590-5 
 
 589-9 
 
 589-3 
 
 590-5 
 
 589-9 
 
 589-3 
 
 588-7 
 
 588-1 
 
 587-5 
 
 588-8 
 
 588-2 
 
 587-6 
 
 587-0 
 
 586-4 
 
 585-8 
 
 587-0 
 
 586-4 
 
 585-8 
 
 585-2 
 
 584-7 
 
 584-1 
 
 585-3 
 
 584-7 
 
 584-1 
 
 583-5 
 
 582-9 
 
 582-3 
 
 583-6 
 
 583-0 
 
 582-4 
 
 581-8 
 
 581-2 
 
 580-6 
 
 581-9 
 
 581-3 
 
 580-7 
 
 580-1 
 
 579-5 
 
 578-9 
 
 580-2 
 
 579-6 
 
 579-0 
 
 578-4 
 
 577-8 
 
 577-3 
 
 578-5 
 
 577-9 
 
 577-3 
 
 576-7 
 
 576-2 
 
 575-6 
 
 576-8 
 
 576-2 
 
 575-6 
 
 575-1 
 
 574-5 
 
 573-9 
 
 575-1 
 
 574-6 
 
 574-0 
 
 573-4 
 
 572-8 
 
 572-2 
 
 573-5 
 
 572-9 
 
 572-3 
 
 571-8 
 
 571-2 
 
 570-6 
 
 571-8 
 
 571-3 
 
 570-7 
 
 570-1 
 
 569-5 
 
 569-0 
 
 570-2 
 
 569-6 
 
 569-0 
 
 568-5 
 
 567-9 
 
 567-3 
 
 568-6 
 
 568-0 
 
 567-4 
 
 566-9 
 
 566-3 
 
 565-7 
 
 566-9 
 
 566-4 
 
 565-8 
 
 565-2 
 
 564-7 
 
 564-1 
 
 565-3 
 
 564-8 
 
 564-2 
 
 563-6 
 
 563-1 
 
 562-5 
 
 563-7 
 
 563-2 
 
 562-6 
 
 562-0 
 
 561-5 
 
 560-9 
 
 562-1 
 
 561-6 
 
 561-0 
 
 560-4 
 
 559-9 
 
 559-3 
 
 560-6 
 
 560-0 
 
 559-4 
 
 558-9 
 
 558-3 
 
 557-7 
 
 559-0 
 
 558-4 
 
 557-9 * 
 
 557-3 
 
 556-7 
 
 556-2 
 
 557-4 
 
 556-9 
 
 556-3 
 
 555-7 
 
 555-2 
 
 554-6 
 
 555-9 
 
 555-3 
 
 554-7 
 
 554-2 
 
 553-6 
 
 553-1 
 
 554-3 
 
 553-8 
 
 553-2 
 
 552-6 
 
 552-1 
 
 551-5 
 
 552-8 
 
 552-2 
 
 551-7 
 
 551-1 
 
 550-6 
 
 550-0 
 
 551-2 
 
 550-7 
 
 550-1 
 
 549-S 
 
 549-0 
 
 548-5 
 
 549-7 
 
 549-2 
 
 548-6 
 
 548-1 
 
 547-5 
 
 547-0 
 
 548-2 
 
 547-7 
 
 547-1 
 
 546-6 
 
 546-0 
 
 545-5 
 
 546-7 
 
 546-2 
 
 545-6 
 
 545-1 
 
 544-5 
 
 544-0 
 
 545-2 
 
 544-7 
 
 544-1 
 
 543-6 
 
 543-0 
 
 542-5 
 
 543-7 
 
 543-2 
 
 542-6 
 
 542-1 
 
 541-5 
 
 541-0 
 
(56-2 
 
 THE ASSAY OF SILVER. 
 
 NITKATE OF 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 i. 
 
 2. 
 
 3. 
 
 4. 
 
 1835 
 
 545-0 
 
 544-4 
 
 543-9 
 
 543-3 
 
 542-8 
 
 1840 
 
 543-5 
 
 542-9 
 
 542-4 
 
 541-8 
 
 541-3 
 
 1845 
 
 542-0 
 
 541-5 
 
 540-9 
 
 540-4 
 
 539-8 
 
 1850 
 
 540-5 
 
 540-0 
 
 539-5 
 
 538-9 
 
 538-4 
 
 1855 
 
 539-1 
 
 538-5 
 
 538-0 
 
 537-5 
 
 536-9 
 
 1860 
 
 537-6 
 
 537-1 
 
 536-6 
 
 536-0 
 
 535-5 
 
 1865 
 
 536-2 
 
 535-7 
 
 535-1 
 
 534-6 
 
 534-0 
 
 1870 
 
 534-8 
 
 534-2 
 
 533-7 
 
 533-2 
 
 532-6 
 
 1875 
 
 533-3 
 
 532-8 
 
 532-3 
 
 531-7 
 
 531-2 
 
 1880 
 
 531-9 
 
 531-4 
 
 530-8 
 
 530-3 
 
 529-8 
 
 1885 
 
 530-5 
 
 530-0 
 
 529-4 
 
 528-9 
 
 528-4 
 
 1890 
 
 529-1 
 
 528-6 
 
 528-0 
 
 527-5 
 
 527-0 
 
 1895 
 
 527-7 
 
 527-2 
 
 526-6 
 
 526-1 
 
 525-6 
 
 1900 
 
 526-3 
 
 525-8 
 
 525-3 
 
 524-7 
 
 524-2 
 
 1905 
 
 524-9 
 
 524-4 
 
 523-9 
 
 523-4 
 
 522-8 
 
 1910 
 
 523-6 
 
 523-0 
 
 522-5 
 
 522-0 
 
 521-5 
 
 1915 
 
 522-2 
 
 521-7 
 
 521-1 
 
 520-6 
 
 520-1 
 
 1920 
 
 520-8 
 
 520-3 
 
 519-8 
 
 519-3 
 
 518-7 
 
 1925 
 
 519-5 
 
 519-0 
 
 518-4 
 
 517-9 
 
 517-4 
 
 1930 
 
 518-1 
 
 517-6 
 
 517-1 
 
 516-6 
 
 516-1 
 
 1935 
 
 516-8 
 
 516-3 
 
 515-8 
 
 515-2 
 
 514-7 
 
 1940 
 
 515-5 
 
 514-9 
 
 514-4 
 
 513-9 
 
 513-4 
 
 1945 
 
 514-1 
 
 513-6 
 
 513-1 
 
 512-6 
 
 512-1 
 
 1950 
 
 512-8 
 
 512-3 
 
 511-8 
 
 511-3 
 
 510-8 
 
 1955 
 
 511-5 
 
 511-0 
 
 510-5 
 
 510-0 
 
 509-5 
 
 1960 
 
 510-2 
 
 509-7 
 
 509-2 
 
 508-7 
 
 508-2 
 
 1965 
 
 508-9 . 
 
 508-4 
 
 507-9 
 
 507-4 
 
 506-9 
 
 1970 
 
 507-6 
 
 507-1 
 
 506-6 
 
 506-1 
 
 505-6 
 
 1975 
 
 506-3 
 
 505-8 
 
 505-3 
 
 504-8 
 
 504-3 
 
 1980 
 
 505-0 
 
 504-5 
 
 504-0 
 
 503-5 
 
 503-0 
 
 1985 
 
 503-8 
 
 503-3 
 
 502-8 
 
 502-3 
 
 501-8 
 
 1990 
 
 502-5 
 
 502-0 
 
 501-5 
 
 501-0 
 
 500-^ 
 
 1995 
 
 501-3 
 
 500-7 
 
 500-2 
 
 499-7 
 
 499-2 
 
 2000 
 
 500-0 
 
 499-5 
 
 499-0 
 
 498-5 
 
 498-0 
 
TABLE FOR THE WET ASSAY OF SILVER. 
 
 663 
 
 
 SILVER continued. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 
 542-2 
 
 541-7 
 
 541-1 
 
 540-6 
 
 540-0 
 
 539-5 
 
 
 540-8 
 
 540-2 
 
 539-7 
 
 539-1 
 
 538-6 
 
 538-0 
 
 
 539-3 
 
 538-7 
 
 538-2 
 
 537-7 
 
 537-1 
 
 536-6 
 
 
 537*8 
 
 537-3 
 
 536-8 
 
 536-2 
 
 535-7 
 
 535-1 
 
 
 536-4 
 
 535-8 
 
 535-3 
 
 534-8 
 
 534-2 
 
 533-7 
 
 
 534-9 
 
 534-4 
 
 533-9 
 
 533-3 
 
 532-8 
 
 532-3 
 
 
 533-5 
 
 533-0 
 
 532-4 
 
 531-9 
 
 531-4 
 
 530-8 
 
 
 532-1 
 
 531-5 
 
 531-0 
 
 530-5 
 
 529-9 
 
 529-4 
 
 
 530-7 
 
 530-1 
 
 529-6 
 
 529-1 
 
 528-5 
 
 528-0 
 
 
 529-3 
 
 528-7 
 
 528-2 
 
 527-7 
 
 527-1 
 
 526-6 
 
 
 527-8 
 
 527-3 
 
 526-8 
 
 526-3 
 
 525-7 
 
 525-2 
 
 
 526-5 
 
 525-9 
 
 525-4 
 
 524-9 
 
 524-3 
 
 523-8 
 
 
 525-1 
 
 524-5 
 
 524-0 
 
 523-5 
 
 523-0 
 
 522-4 
 
 
 523-7 
 
 523-2 
 
 522-6 
 
 522-1 
 
 521-6 
 
 521-0 
 
 
 522-3 
 
 521-8 
 
 521-3 
 
 520-7 
 
 520-2 
 
 519-7 
 
 
 520-9 
 
 520-4 
 
 519-9 
 
 519-4 
 
 518-8 
 
 518-3 
 
 
 519-6 
 
 519-1 
 
 518-5 
 
 518-0 
 
 517-5 
 
 517-0 
 
 
 518-2 
 
 517-7 
 
 517-2 
 
 516-7 
 
 516-1 
 
 515-6 
 
 
 516-9 
 
 516-4 
 
 515-8 
 
 515-3 
 
 514-8 
 
 514-3 
 
 
 515-5 
 
 515-0 
 
 514-5 
 
 514-0 
 
 513-5 
 
 512-9 
 
 
 514-2 
 
 513-7 
 
 513-2 
 
 512-7 
 
 512-1 
 
 511-6 
 
 
 512-9 
 
 512-4 
 
 511-9 
 
 511-3 
 
 510-8 
 
 510-3 
 
 
 511-6 
 
 511-0 
 
 510-5 
 
 510-0 
 
 509-5 
 
 509-0 
 
 
 510-3 
 
 509-7 
 
 509-2 
 
 508-7 
 
 508-2 
 
 507-7 
 
 
 508-9 
 
 508-4 
 
 507-9 
 
 507-4 
 
 506-9 
 
 506-4 
 
 
 507-6 
 
 507-1 
 
 506-6 
 
 506-1 
 
 505-6 
 
 505-1 
 
 
 506-4 
 
 505-8 
 
 505-3 
 
 504-8 
 
 504-3 
 
 503-8 
 
 
 505-1 
 
 504-6 
 
 504-1 
 
 503-5 
 
 503-0 
 
 502-5 
 
 
 503-8 
 
 503-3 
 
 502-8 
 
 502-3 
 
 501-8 
 
 501-3 
 
 
 502-5 
 
 509-0 
 
 501-5 
 
 501-0 
 
 500-5 
 
 500-0 
 
 
 501-3 
 
 500-8 
 
 500-2 
 
 499-7 
 
 499-2 
 
 498-7 
 
 
 500-0 
 
 499-5 
 
 499-0 
 
 498-5 
 
 498-0 
 
 497-5 
 
 
 498-7 
 
 498-2 
 
 497-7 
 
 497-2 
 
 496-7 
 
 496-2 
 
 
 497-5 
 
 497-0 
 
 496-5 
 
 496-0 
 
 495-5 
 
 495-0 
 
C64 
 
 THE ASSAY OF SILVER. 
 
 Tables for Determining the Standard of any Silver 
 approximatively containing 
 
 COMMON 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 i. 
 
 2. 
 
 3. 
 
 4. 
 
 
 1000 
 
 1000-0 
 
 
 
 
 
 
 1005 
 
 995-0 
 
 996-0 
 
 997-0 
 
 998-0 
 
 999-0 
 
 
 1010 
 
 990-1 
 
 991-1 
 
 992-1 
 
 993-1 
 
 994-1 
 
 
 1015 
 
 985-2 
 
 986-2 
 
 987-2 
 
 988-2 
 
 989-2 
 
 
 1020 
 
 980-4 
 
 981-4 
 
 982-4 
 
 983-3 
 
 984-3 
 
 
 1025 
 
 975-6 
 
 976-6 
 
 977-6 
 
 978-5 
 
 979-5 
 
 
 1030 
 
 970-9 
 
 971-8 
 
 972-8 
 
 973-8 
 
 974-8 
 
 
 1035 
 
 966-2 
 
 967-1 
 
 968-1 
 
 969-1 
 
 970-0 
 
 
 1040 
 
 961-5 
 
 962-5 
 
 963-5 
 
 964-4 
 
 965-4 
 
 
 1045 
 
 956-9 
 
 957-9 
 
 958-8 
 
 959-8 
 
 960-8 
 
 
 1050 
 
 952-4 
 
 953-3 
 
 954-3 
 
 955-2 
 
 956-2 
 
 
 1055 
 
 947-9 
 
 948-8 
 
 949-8 
 
 950-7 
 
 951-7 
 
 
 1060 
 
 943-4 
 
 944-3 
 
 945-3 
 
 946-2 
 
 947-2 
 
 
 1065 
 
 939-0 
 
 939-9 
 
 940-8 
 
 941-8 
 
 942-7 
 
 
 1070 
 
 934-6 
 
 935-5 
 
 936-4 
 
 937-4 
 
 938-3 
 
 
 1075 
 
 930-2 
 
 931-2 
 
 932-1 
 
 933-0 
 
 933-9 
 
 
 1080 
 
 925-9 
 
 926-8 
 
 927-8 
 
 928-7 
 
 929-6 
 
 
 1085 
 
 921-7 
 
 922-6 
 
 923-5 
 
 924-4 
 
 925-3 
 
 
 1090 
 
 917-4 
 
 918-3 
 
 919-3 
 
 920-2 
 
 921-1 
 
 
 1095 
 
 913-2 
 
 914-2 
 
 915-1 
 
 916-0 
 
 917-0 
 
 
 1100 
 
 909-1 
 
 910-0 
 
 910-9 
 
 911-8 
 
 912-7 
 
 
 1105 
 
 905-0 
 
 905-9 
 
 906-8 
 
 907-7 
 
 908-6 
 
 
 1110 
 
 900-9 
 
 901-8 
 
 902-7 
 
 903-6 
 
 904-5 
 
 
 1115 
 
 896-9 
 
 897-8 
 
 898-6 
 
 899-5 
 
 900-4 
 
 
 1120 
 
 892-9 
 
 893-7 
 
 894-6 
 
 895-5 
 
 896-4 
 
 
 1125 
 
 888-9 
 
 889-8 
 
 890-7 
 
 891-6 
 
 892-4 
 
 
 1130 
 
 885-0 
 
 885-8 
 
 886-7 
 
 887-6 
 
 888-5 
 
 
 1135 
 
 881-1 
 
 881-9 
 
 882-8 
 
 883-7 
 
 884-6 
 
 
 1140 
 
 877-2 
 
 878-1 
 
 878-9 
 
 879-8 
 
 880-7 
 
 
 1145 
 
 873-4 
 
 874-2 
 
 875-1 
 
 876-0 
 
 876-9 
 
 
 1150 
 
 869-6 
 
 870-4 
 
 871-3 
 
 872-2 
 
 873-0 
 
 
 1155 
 
 865-8 
 
 866-7 
 
 867-5 
 
 868-4 
 
 869-3 
 
 
 1160 
 
 862-1 
 
 862-9 
 
 863-8 
 
 864-7 
 
 865-5 
 
 
 1165 
 
 858-4 
 
 859-2 
 
 860-1 
 
 860-9 
 
 861-8 
 
 
 1170 
 
 854-7 
 
 855-6 
 
 856-4 
 
 857-3 
 
 858-1 
 
 
 1175 
 
 851-1 
 
 851-9 
 
 852-8 
 
 853-6 
 
 854-5 
 
 
 1180 
 
 847-5 
 
 848-3 
 
 849-2 
 
 850-0 
 
 850-8 
 
 
 1185 
 
 843-9 
 
 844-7 
 
 845-6 
 
 846-4 
 
 847-3 
 
 
TABLE FOB THE WET ASSAY OF SILVER. 
 
 665 
 
 Alloy by employing an Amount of Alloy always 
 the same Amount of Silver. 
 
 SALT. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 1000-0 
 
 
 
 
 
 
 995-0 
 
 996-0 
 
 997-0 
 
 998-0 
 
 999-0 
 
 1000-0 
 
 990-1 
 
 991-1 
 
 992-1 
 
 993-1 
 
 994-1 
 
 995-1 
 
 985-3 
 
 986-3 
 
 987-2 
 
 988-2 
 
 989-2 
 
 990-2 
 
 980-5 
 
 981-5 
 
 982-4 
 
 983-4 
 
 984-4 
 
 985-4 
 
 975-7 
 
 976-7 
 
 977-7 
 
 978-6 
 
 979-6 
 
 980-6 
 
 971-0 
 
 972-0 
 
 972-9 
 
 973-9 
 
 974-9 
 
 975-8 
 
 966-3 
 
 967-3 
 
 968-3 
 
 969-2 
 
 970-2 
 
 971-1 
 
 961-7 
 
 962-7 
 
 963-6 
 
 964-6 
 
 965-5 
 
 966-5 
 
 957-1 
 
 958-1 
 
 959-0 
 
 960-0 
 
 960-9 
 
 961-9 
 
 952-6 
 
 953-5 
 
 954-5 
 
 955-4 
 
 956-4 
 
 957-3 
 
 948-1 
 
 949-1 
 
 950-0 
 
 950-9 
 
 951-9 
 
 952-8 
 
 943-7 
 
 944-6 
 
 945-5 
 
 946-5 
 
 947-4 
 
 948-4 
 
 939-3 
 
 940-2 
 
 941-1 
 
 942-1 
 
 943-0 
 
 943-9 
 
 934-9 
 
 935-8 
 
 936-7 
 
 937-7 
 
 938-6 
 
 939-5 
 
 930-6 
 
 931-5 
 
 932-4 
 
 933-3 
 
 934-3 
 
 935-2 
 
 926-3 
 
 927-2 
 
 928-1 
 
 929-0 
 
 930-0 
 
 930-9 
 
 922-0 
 
 922-9 
 
 923-8 
 
 924-8 
 
 925-7 
 
 926-6 
 
 917-8 
 
 918-7 
 
 919-6 
 
 920-5 
 
 921-5 
 
 922-4 
 
 913-6 
 
 914-5 
 
 915-4 
 
 916-4 
 
 917-3 
 
 918-2 
 
 909-5 
 
 910-4 
 
 911-3 
 
 912-2 
 
 913-1 
 
 914-0 
 
 905-4 
 
 906-3 
 
 907-2 
 
 908-1 
 
 909-0 
 
 909-9 
 
 901-3 
 
 902-2 
 
 903-1 
 
 904-0 
 
 904-9 
 
 905-8 
 
 897-3 
 
 898-2 
 
 899-1 
 
 900-0 
 
 900-9 
 
 901-8 
 
 893-3 
 
 894-2 
 
 895-1 
 
 896-0 
 
 896-9 
 
 897-8 
 
 889-4 
 
 890-3 
 
 891-1 
 
 892-0 
 
 892-9 
 
 893-8 
 
 885-5 
 
 886-3 
 
 887-2 
 
 888-1 
 
 889-0 
 
 889-9 
 
 881-6 
 
 882-5 
 
 883-3 
 
 884-2 
 
 885-1 
 
 886-0 
 
 877-7 
 
 878-6 
 
 879-5 
 
 880-3 
 
 881-2 
 
 882-1 
 
 873-9 
 
 874-8 
 
 875-7 
 
 876-5 
 
 877-4 
 
 878-3 
 
 870-1 
 
 871-0 
 
 871-9 
 
 872-7 
 
 873-6 
 
 874-5 
 
 866-4 
 
 867-2 
 
 868-1 
 
 869-0 
 
 869-8 
 
 870-7 
 
 862-7 
 
 863-5 
 
 864-4 
 
 865-2 
 
 866-1 
 
 866-9 
 
 859-0 
 
 859-8 
 
 860-7 
 
 861-5 
 
 862-4 
 
 863-2 
 
 855-3 
 
 856-2 
 
 857-0 
 
 857-9 
 
 858-7 
 
 859-6 
 
 851-7 
 
 852-5 
 
 853-4 
 
 854-2 
 
 855-1 
 
 855-9 
 
 848-1 
 
 848-9 
 
 849-8 
 
 850-6 
 
 851-5 
 
 852-3 
 
6G6 
 
 THE ASSAY OF SILVER. 
 
 COMMON 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 i. 
 
 2. 
 
 3. 
 
 4. 
 
 1190 
 
 840-3 
 
 841-2 
 
 842-0 
 
 842-9 
 
 843-7 
 
 1195 
 
 836-8 
 
 837-7 
 
 838-5 
 
 839-3 
 
 840-2 
 
 1200 
 
 833-3 
 
 834-2 
 
 835-0 
 
 835-8 
 
 836-7 
 
 1205 
 
 829-9 
 
 830-7 
 
 831-5 
 
 832-4 
 
 833-2 
 
 1210 
 
 826-4 
 
 827-3 
 
 828-1 
 
 828-9 
 
 829-7 
 
 1215 
 
 823-0 
 
 823-9 
 
 824-7 
 
 825-5 
 
 826-3 
 
 1220 
 
 819-7 
 
 820-5 
 
 821-3 
 
 822-1 
 
 822-9 
 
 1225 
 
 816-3 
 
 817-1 
 
 818-0 
 
 818-8 
 
 819-6 
 
 1230 
 
 813-0 
 
 813-8 
 
 814-6 
 
 815-4 
 
 816-3 
 
 1235 
 
 809-7 
 
 810-5 
 
 811-3 
 
 812-1 
 
 813-0 
 
 1240 
 
 806-5 
 
 807-3 
 
 808-1 
 
 808-9 
 
 809-7 
 
 1245 
 
 803-2 
 
 804-0 
 
 804-8 
 
 805-6 
 
 806-4 
 
 1250 
 
 800-0 
 
 800-8 
 
 801-6 
 
 802-4 
 
 803-2 
 
 1255 
 
 796-8 
 
 797-6 
 
 798-4 
 
 799-2 
 
 800-0 
 
 1260 
 
 793-6 
 
 794-4 
 
 795-2 
 
 796-0 
 
 796-8 
 
 1265 
 
 790-5 
 
 791-3 
 
 792-1 
 
 792-9 
 
 793-7 
 
 1270 
 
 787-4 
 
 788-2 
 
 789-0 
 
 789-8 
 
 790-5 
 
 1275 
 
 784-3 
 
 785-1 
 
 785-9 
 
 786-7 
 
 787-4 
 
 1280 
 
 781-2 
 
 782-0 
 
 782-8 
 
 783-6 
 
 784-4 
 
 1285 
 
 778-2 
 
 779-0 
 
 779-8 
 
 780-5 
 
 781-3 
 
 1290 
 
 . 775-2 
 
 776-0 
 
 776-7 
 
 777-5 
 
 778-3 
 
 1295 
 
 772-2 
 
 773-0 
 
 773-7 
 
 774-5 
 
 775-3 
 
 1300 
 
 769-2 
 
 770-0 
 
 770-8 
 
 771-5 
 
 772-3 
 
 1305 
 
 766-3 
 
 767-0 
 
 767-8 
 
 768-6 
 
 769-3 
 
 1310 
 
 763-4 
 
 764-1 
 
 764-9 
 
 765-6 
 
 766-4 
 
 1315 
 
 760-5 
 
 761-2 
 
 762-0 
 
 762-7 
 
 763-5 
 
 1320 
 
 757-6 
 
 758-3 
 
 759-1 
 
 759-8 
 
 760-6 
 
 1325 
 
 754-7 
 
 755-5 
 
 756-2 
 
 757-0 
 
 757-7 
 
 1330 
 
 751-9 
 
 752-6 
 
 753-4 
 
 754-1 
 
 754-9 
 
 1335 
 
 749-1 
 
 749-8 
 
 750-6 
 
 751-3 
 
 752-1 
 
 1340 
 
 746-3 
 
 747-0 
 
 747-8 
 
 748-5 
 
 749-2 
 
 1345 
 
 743-5 
 
 744-2 
 
 745-0 
 
 745-7 
 
 746-5 
 
 1350 
 
 740-7 
 
 741-5 
 
 742-2 
 
 743-0 
 
 743-7 
 
 1355 
 
 738-0 
 
 738-7 
 
 739-5 
 
 740-2 
 
 741-0 
 
 1360 
 
 735-3 
 
 736-0 
 
 736-8 
 
 737-5 
 
 738-2 
 
 1365 
 
 732-6 
 
 733-3 
 
 734-1 
 
 734-8 
 
 735-5 
 
 1370 
 
 729-9 
 
 730-7 
 
 731-4 
 
 732-1 
 
 732-8 
 
 1375 
 
 727-3 
 
 728-0 
 
 728-7 
 
 729-4 
 
 730-2 
 
 1380 
 
 724-6 
 
 725-4 
 
 726-1 
 
 726-8 
 
 727-5 
 
 1385 
 
 722-0 
 
 722-7 
 
 723-5 
 
 724-2 
 
 724-9 
 
 1390 
 
 719-4 
 
 720-1 
 
 720-9 
 
 721-6 
 
 722-3 
 
 1395 
 
 716-8 
 
 717-6 
 
 718-3 
 
 719-0 
 
 719-7 
 
 1400 
 
 714-3 
 
 715-0 
 
 715-7 
 
 716-4 
 
 717-1 
 
TABLE FOE THE WET ASSAY OF SILVEE. 
 
 
 SALT. continued. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 
 844-5 
 
 845-4 
 
 846-2 
 
 847-1 
 
 847-9 
 
 848-7 
 
 
 841-0 
 
 841-8 
 
 842-7 
 
 843-5 
 
 844-3 
 
 845-2 
 
 
 837-5 
 
 838-3 
 
 839-2 
 
 840-0 
 
 840-8 
 
 841-7 
 
 
 834-0 
 
 834-8 
 
 835-7 
 
 836-5 
 
 837-3 
 
 838-2 
 
 
 830-6 
 
 831-4 
 
 832-2 
 
 833-1 
 
 833-9 
 
 834-7 
 
 
 827-2 
 
 828-0 
 
 828-8 
 
 829-6 
 
 830-4 
 
 831-3 
 
 
 823-8 
 
 824-6 
 
 825-4 
 
 826-2 
 
 827-0 
 
 827-9 
 
 
 820-4 
 
 821-2 
 
 822-0 
 
 822-9 
 
 823-7 
 
 824-5 
 
 
 817-1 
 
 817-9 
 
 818-7 
 
 819-5 
 
 820-3 
 
 821-1 
 
 
 813-8 
 
 814-6 
 
 815-4 
 
 816-2 
 
 817-0 
 
 817-8 
 
 
 810-5 
 
 811-3 
 
 812-1 
 
 812-9 
 
 813-7 
 
 814-5 
 
 
 807-2 
 
 808-0 
 
 808-8 
 
 809-6 
 
 810-4 
 
 811-2 
 
 
 804-0 
 
 804-8 
 
 805-6 
 
 806-4 
 
 807-2 
 
 808-0 
 
 
 800-8 
 
 801-6 
 
 802-4 
 
 803-2 
 
 804-0 
 
 804-8 
 
 
 797-6 
 
 798-4 
 
 799-2 
 
 800-0 
 
 800-8 
 
 801-6 
 
 
 794-5 
 
 795-3 
 
 796-0 
 
 796-8 
 
 797-6 
 
 798-4 
 
 
 791-3 
 
 792-1 
 
 792-9 
 
 793-7 
 
 794-5 
 
 795-3 
 
 
 788-2 
 
 789-0 
 
 789-8 
 
 790-6 
 
 791-4 
 
 792-2 
 
 
 785-2 
 
 786-0 
 
 786-7 
 
 787-5 
 
 788-3 
 
 789-1 
 
 
 782-1 
 
 782-9 
 
 783-7 
 
 784-4 
 
 785-2 
 
 786-0 
 
 
 779-1 
 
 779-8 
 
 780-6 
 
 781-4 
 
 782-2 
 
 782-9 
 
 
 776-1 
 
 776-8 
 
 777-6- 
 
 778-4 
 
 779-1 
 
 779-9 
 
 
 773-1 
 
 773-8 
 
 774-6 
 
 775-4 
 
 776-1 
 
 776-9 
 
 
 770-1 
 
 770-9 
 
 771-6 
 
 772-4 
 
 773-2 
 
 773-9 
 
 
 767-2 
 
 767-9 
 
 768-7 
 
 769-5 
 
 770-2 
 
 771-0 
 
 
 764-3 
 
 765-0 
 
 765-8 
 
 766-5 
 
 767-3 
 
 768-1 
 
 
 761-4 
 
 762-1 
 
 762-9 
 
 763-6 
 
 764-4 
 
 765-2 
 
 
 758-5 
 
 759-2 
 
 760-0 
 
 760-7 
 
 761-5 
 
 762-3 
 
 
 755-6 
 
 756-4 
 
 757-1 
 
 757-9 
 
 758-6 
 
 759-4 
 
 
 752-8 
 
 753-6 
 
 754-3 
 
 755-1 
 
 755-8 
 
 756-6 
 
 
 750-0 
 
 750-7 
 
 751-5 
 
 752-2 
 
 753-0 
 
 753-7 
 
 
 747-2 
 
 748-0 
 
 748-7 
 
 749-4 
 
 750-2 
 
 750-9 
 
 
 744-4 
 
 745-2 
 
 745-9 
 
 746-7 
 
 747-4 
 
 748-1 
 
 
 741-7 
 
 742-4 
 
 743-2 
 
 743-9 
 
 744-6 
 
 745-4 
 
 
 739-0 
 
 739-7 
 
 740-4 
 
 741-2 
 
 741-9 
 
 742-6 
 
 
 736-3 
 
 737-0 
 
 737-7 
 
 738-5 
 
 739-2 
 
 739-9 
 
 
 733-6 
 
 734-3 
 
 735-0 
 
 735-8 
 
 736-5 
 
 737-2 
 
 
 730-9 
 
 731-6 
 
 732-4 
 
 733-2 
 
 733-8 
 
 734-5 
 
 
 728-3 
 
 729-0 
 
 729-7 
 
 730-4 
 
 731-2 
 
 731-9 
 
 
 725-6 
 
 726-3 
 
 727-1 
 
 727-8 
 
 728-5 
 
 729-2 
 
 
 723-0 
 
 723-7 
 
 724-5 
 
 725-2 
 
 725-9 
 
 726-6 
 
 
 720-4 
 
 721-1 
 
 721-9 
 
 722-6 
 
 723-3 
 
 724-0 
 
 
 717-9 
 
 718-6 
 
 719-3 
 
 720-0 
 
 720-7 
 
 721-4 
 
668 
 
 THE ASSAY OF SILVER. 
 
 COMMON 
 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 1. 
 
 2. 
 
 3. 
 
 4. 
 
 1405 
 
 711*7 
 
 712-5 
 
 713-2 
 
 713-9 
 
 714-6 
 
 
 1410 
 
 709-2 
 
 709-9 
 
 710-6 
 
 711-3 
 
 712-1 
 
 
 1415 
 
 706-7 
 
 707-4 
 
 708-1 
 
 708-8 
 
 709-5 
 
 
 1420 
 
 704-2 
 
 704-9 
 
 705-6 
 
 706-3 
 
 707-0 
 
 
 1425 
 
 701-8 
 
 702-5 
 
 703-2 
 
 703-9 
 
 704-6 
 
 
 1430 
 
 699-3 
 
 700-0 
 
 700-7 
 
 701-4 
 
 702-1 
 
 
 1435 
 
 696-9 
 
 697-6 
 
 698-3 
 
 698-9 
 
 699-6 
 
 
 1440 
 
 694-4 
 
 695-1 
 
 695-8 
 
 696-5 
 
 697-2 
 
 
 1445 
 
 692-0 
 
 692-7 
 
 693-4 
 
 694-1 
 
 694-8 
 
 
 1450 
 
 689-7 
 
 690-3 
 
 691-0 
 
 691-7 
 
 692-4 
 
 
 1455 
 
 687-3 
 
 688-0 
 
 688-7 
 
 689-3 
 
 690-0 
 
 
 1460 
 
 684-9 
 
 685-6 
 
 686-3 
 
 687-0 
 
 687-7 
 
 
 1465 
 
 682-6 
 
 683-3 
 
 684-0 
 
 684-6 
 
 685-3 
 
 
 1470 
 
 680-3 
 
 680-9 
 
 681-6 
 
 682-3 
 
 683-0 
 
 
 1475 
 
 678-0 
 
 678-6 
 
 679-3 
 
 680-0 
 
 680-7 
 
 
 1480 
 
 675-7 
 
 676-3 
 
 677-0 
 
 677-7 
 
 678-4 
 
 
 1485 
 
 673-4 
 
 674-1 
 
 674-7 
 
 675-4 
 
 676-1 
 
 
 1490 
 
 671-1 
 
 671-8 
 
 672-5 
 
 673-1 
 
 673-8 
 
 
 1495 
 
 668-9 
 
 669-6 
 
 670-2 
 
 670-9 
 
 671-6 
 
 
 1500 
 
 666-7 
 
 667-3 
 
 668-0 
 
 668-7 
 
 669-3 
 
 
 1505 
 
 664-5 
 
 665-1 
 
 665-8 
 
 666-4 
 
 667-1 
 
 
 1510 
 
 662-3 
 
 662-9 
 
 663-6 
 
 664-2 
 
 664-9 
 
 
 1515 
 
 660-1 
 
 660-7 
 
 661-4 
 
 662-0 
 
 662-7 
 
 
 1520 
 
 657-9 
 
 658-5 
 
 659-2 
 
 659-9 
 
 660-5 
 
 
 1525 
 
 655-7 
 
 656-4 
 
 657-0 
 
 657-7 
 
 658-4 
 
 
 1530 
 
 653-6 
 
 654-2 
 
 654-9 
 
 655-6 
 
 656-2 
 
 
 1535 
 
 651-5 
 
 652-1 
 
 652-8 
 
 653-4 
 
 654-1 
 
 
 1540 
 
 649-4 
 
 650-0 
 
 650-6 
 
 651-3 
 
 651-9 
 
 
 1545 
 
 647-2 
 
 647-9 
 
 648-5 
 
 649-2 
 
 649-8 
 
 
 1550 
 
 645-2 
 
 645-8 
 
 646-4 
 
 647-1 
 
 647-7 
 
 
 1555 
 
 643-1 
 
 643-7 
 
 644-4 
 
 645-0 
 
 645-7 
 
 
 1560 
 
 641-0 
 
 641-7 
 
 642-3 
 
 642-9 
 
 643-6 
 
 
 1565 
 
 639-0 
 
 639-6 
 
 640-3 
 
 640-9 
 
 641-5 
 
 
 1570 
 
 636-9 
 
 637-6 
 
 638-2 
 
 638-8 
 
 639-5 
 
 
 1575 
 
 634-9 
 
 635-6 
 
 636-2 
 
 636-8 
 
 637-5 
 
 
 1580 
 
 632-9 
 
 633-5 
 
 634-2 
 
 634-8 
 
 635-4 
 
 
 1585 
 
 630-9 
 
 631-5 
 
 632-2 
 
 632-8 
 
 633-4 
 
 
 1590 
 
 628-9 
 
 629-6 
 
 630-2 
 
 630-8 
 
 631-4 
 
 
 1595 
 
 627-0 
 
 627-6 
 
 628-2 
 
 628-8 
 
 629-5 
 
 
 1600 
 
 625-0 
 
 625-6 
 
 626-2 
 
 626-9 
 
 627-5 
 
 
 1605 
 
 623-1 
 
 623-7 
 
 624-3 
 
 624-9 
 
 625-5 
 
 
 1610 
 
 621-1 
 
 621-7 
 
 622-4 
 
 623-0 
 
 623-6 
 
 
 1615 
 
 619-2 
 
 619-8 
 
 620-4 
 
 621-0 
 
 621-7 
 
 
TABLE FOR THE WET ASSAY OF SILVER. 
 
 SALT continued. 
 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 
 715-3 
 
 716-0 716-7 
 
 717-4 
 
 718-1 
 
 718-9 
 
 
 712-8 
 
 713-5 714-2 
 
 714-9 
 
 715-6 
 
 716-3 
 
 
 710-2 
 
 710-9 
 
 711-7 
 
 712-4 
 
 713-1 
 
 713-8 
 
 
 707-7 
 
 708-4 
 
 709-2 
 
 709-9 
 
 710-6 
 
 711-3 
 
 
 705-3 
 
 706-0 
 
 706-7 
 
 707-4 
 
 708-1 
 
 708-8 
 
 
 702-8 
 
 703-5 
 
 704-2 
 
 704-9 
 
 705-6 
 
 706-3 
 
 
 700-3 
 
 701-0 
 
 701-7 
 
 702-4 
 
 703-1 
 
 703-8 
 
 
 697-9 
 
 698-6 
 
 699-3 
 
 700-0 
 
 700-7 
 
 701-4 
 
 
 695-5 
 
 696-2 
 
 696-9 
 
 697-6 
 
 698-3 
 
 699-0 
 
 
 693-1 
 
 693-8 
 
 694-5 
 
 695-2 
 
 695-9 
 
 696-6 
 
 
 690-7 
 
 691-4 
 
 692-1 
 
 692-8 
 
 693-5 
 
 694-2 
 
 
 688-4 
 
 689-0 
 
 689-7 
 
 690-4 
 
 691-1 
 
 691-8 
 
 
 686-0 
 
 686-7 
 
 687-4 
 
 688-0 
 
 688-7 
 
 689-4 
 
 
 683-7 
 
 684-3 
 
 685-0 
 
 685-7 
 
 686-4 
 
 687-1 
 
 
 681-4 
 
 682-0 
 
 682-7 
 
 683-4 
 
 684-1 
 
 684-7 
 
 
 679-1 
 
 679-7 
 
 680-4 
 
 681-1 
 
 681-8 
 
 682-4 
 
 
 676-8 
 
 677-4 
 
 678-1 
 
 678-8 
 
 679-5 
 
 680-1 
 
 
 674-5 
 
 675-2 
 
 675-8 
 
 676-5 
 
 677-2 
 
 677-8 
 
 
 672-2 
 
 672-9 
 
 673-6 
 
 674-2 
 
 674-9 
 
 675-6 
 
 
 670-0 
 
 670-7 
 
 671-3 
 
 672-0 
 
 672-7 
 
 673-3 
 
 
 667-8 
 
 668-4 
 
 669-1 
 
 669-8 
 
 670-4 
 
 671-1 
 
 
 665-6 
 
 666-2 
 
 666-9 
 
 667-5 
 
 668-2 
 
 668-9 
 
 
 663-4 
 
 664-0 
 
 664-7 
 
 665-3 
 
 666-0 
 
 666-7 
 
 
 661-2 
 
 661-8 
 
 662-5 
 
 663-2 
 
 663-8 
 
 664-5 
 
 
 659-0 
 
 659-7 
 
 660-3 
 
 661-0 
 
 661-6 
 
 662-3 
 
 
 656-9 
 
 657-5 
 
 658-2 
 
 658-8 
 
 65?-5 
 
 660-1 
 
 
 654-7 
 
 655-4 
 
 656-0 
 
 656-7 
 
 657-3 
 
 658-0 
 
 
 652-6 
 
 653-2 
 
 653-9 
 
 654-5 
 
 655-2 
 
 655-8 
 
 
 650-5 
 
 651-1 
 
 651-8 
 
 652-4 
 
 653-1 
 
 653-7 
 
 
 648-4 
 
 649-0 
 
 649-7 
 
 650-3 
 
 651-0 
 
 651-6 
 
 
 646-3 
 
 646-9 
 
 647-6 
 
 648-2 
 
 648-9 
 
 649-5 
 
 
 644-2 
 
 644-9 
 
 645-5 
 
 646-1 
 
 646-8 
 
 647-4 
 
 
 642-2 
 
 642-8 
 
 643-4 
 
 644-1 
 
 644-7 
 
 645-4 
 
 
 640-1 
 
 640-8 
 
 641-4 
 
 642-0 
 
 642-7 
 
 643-3 
 
 
 638-1 
 
 638-7 
 
 639-4 
 
 640-0 
 
 640-6 
 
 641-3 
 
 
 636-1 
 
 636-7 
 
 637-3 
 
 638-0 
 
 638-6 
 
 639-2 
 
 
 634-1 
 
 634-7 
 
 635-3 
 
 636-0 
 
 636-6 
 
 637-2 
 
 
 632-1 
 
 632-7 
 
 633-3 
 
 634-0 
 
 634-6 
 
 635-2 
 
 
 630-1 
 
 630-7 
 
 631-3 
 
 632-0 
 
 632-6 
 
 633-2 
 
 
 628-1 
 
 628-7 
 
 629-4 
 
 630-0 
 
 630-6 
 
 631-2 
 
 
 626-2 
 
 626-8 
 
 627-4 
 
 628-0 
 
 628-7 
 
 629-3 
 
 
 624-2 
 
 624-8 
 
 625-5 
 
 626-1 
 
 626-7 
 
 627-3 
 
 
 622-3 
 
 622-9 
 
 623-5 
 
 624-1 
 
 624-8 
 
 625-4 
 
670 
 
 THE ASSAY OF SILVER. 
 
 COMMON 
 
 Weight of 
 
 
 
 
 
 
 
 Assay in 
 
 0. 
 
 i. 
 
 2. 
 
 3. 
 
 4. 
 
 
 Milligrs. 
 
 
 
 
 
 
 
 1620 
 
 617-3 
 
 617-9 
 
 618-5 
 
 619-1 
 
 619-7 
 
 
 1625 
 
 615-4 
 
 616-0 
 
 616-6 
 
 617-2 
 
 617-8 
 
 
 1630 
 
 613-5 
 
 614-1 
 
 614-7 
 
 615-3 
 
 615-9 
 
 
 1635 
 
 611-6 
 
 612-2 
 
 612-8 
 
 613-5 
 
 614-1 
 
 
 1640 
 
 609-8 
 
 610-4 
 
 611-0 
 
 611-6 
 
 612-2 
 
 
 1645 
 
 607-9 
 
 608-5 
 
 609-1 
 
 609-7 
 
 610-3 
 
 
 1650 
 
 606-1 
 
 606-7 
 
 607-3 
 
 607-9 
 
 608-5 
 
 
 1655 
 
 604-2 
 
 604-8 
 
 605-4 
 
 606-0 
 
 606-6 
 
 
 1660 
 
 602-4 
 
 603-0 
 
 603-6 
 
 604-2 
 
 604-8 
 
 
 1665 
 
 600-6 
 
 601-2 
 
 601-8 
 
 602-4 
 
 603-0 
 
 
 1670 
 
 598-8 
 
 699-4 
 
 600-0 
 
 600-6 
 
 601-2 
 
 
 1675 
 
 597-0 
 
 597-6 
 
 598-2 
 
 598-8 
 
 599-4 
 
 
 1680 
 
 595-2 
 
 595-8 
 
 596-4 
 
 597-0 
 
 597-6 
 
 
 1685 
 
 593-5 
 
 594-1 
 
 594-7 
 
 595-2 
 
 595-8 
 
 
 1690 
 
 591-7 
 
 592-3 
 
 o92-9 
 
 593-5 
 
 594-1 
 
 
 1695 
 
 590-0 
 
 590-6 
 
 591-1 
 
 591-7 
 
 592-3 
 
 
 1700 
 
 588-2 
 
 588-8 
 
 589-4 
 
 590-0 
 
 590-6 
 
 
 1705 
 
 586-5 
 
 587-1 
 
 587-7 
 
 588-3 
 
 588-9 
 
 
 1710 
 
 584-8 
 
 585-4 
 
 586-0 
 
 586-5 
 
 587-1 
 
 
 1715 
 
 583-1 
 
 583-7 
 
 584-3 
 
 584-8 
 
 585-4 
 
 
 1720 
 
 581-4 
 
 582-0 
 
 582-6 
 
 583-1 
 
 583-7 
 
 
 1725 
 
 579-7 
 
 580-3 
 
 580-9 
 
 581-4 
 
 582-0 
 
 
 1730 
 
 578-0 
 
 578-6 
 
 579-2 
 
 579-8 
 
 580-3 
 
 
 1735 
 
 576-4 
 
 576-9 
 
 577-5 
 
 578-1 
 
 578-7 
 
 
 1740 
 
 574-7 
 
 575-3 
 
 575-9 
 
 576-4 
 
 577-0 
 
 
 1745 
 
 573-1 
 
 573-6 
 
 574-2 
 
 574-8 
 
 575-4 
 
 
 1750 
 
 571-4 
 
 572-0 
 
 572-6 
 
 573-1 
 
 573-7 
 
 
 1755 
 
 569-8 
 
 570-4 
 
 570-9 
 
 571-5 
 
 572-1 
 
 
 1760 
 
 568-2 
 
 568-7 
 
 569-3 
 
 569-9 
 
 570-4 
 
 
 1765 
 
 566-6 
 
 567-1 
 
 567-7 
 
 568-3 
 
 568-8 
 
 
 1770 
 
 565-0 
 
 565-5 
 
 566-1 
 
 566-7 
 
 567-2 
 
 
 1775 
 
 563-4 
 
 563-9 
 
 564-5 
 
 565-1 
 
 565-6 
 
 
 1780 
 
 561-8 
 
 562-4 
 
 562-9 
 
 563-5 
 
 564-0 
 
 
 1785 
 
 560-2 
 
 560-8 
 
 561-3 
 
 561-9 
 
 562-5 
 
 
 1790 
 
 558-7 
 
 559-2 
 
 559-8 
 
 560-3 
 
 560-9 
 
 
 1795 
 
 557-1 
 
 557-7 
 
 558-2 
 
 558-8 
 
 559-3 
 
 
 1800 
 
 555-6 
 
 556-1 
 
 556-7 
 
 557-2 
 
 557-8 
 
 
 1805 
 
 554-0 
 
 554-6 
 
 555-1 
 
 555-7 
 
 556-2 
 
 
 1810 
 
 552-5 
 
 553-0 
 
 553-6 
 
 554-1 
 
 554-7 
 
 
 1815 
 
 551-0 
 
 551-5 
 
 552-1 
 
 552-6 
 
 553-2 
 
 
 1820 
 
 549-4 
 
 550-0 
 
 550-5 
 
 551-1 
 
 551-6 
 
 
 1825 
 
 547-9 
 
 548-5 
 
 549-0 
 
 549-6 
 
 550-1 
 
 
 1830 
 
 546-4 
 
 547-0 
 
 547-5 
 
 548-1 
 
 548-6 
 
 
TABLE FOR THE WET ASSAY OF SILVER. 
 
 071 
 
 I 
 
 SALT continued. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 
 10. 
 
 
 620-4 
 
 621-0 
 
 621-6 
 
 622-2 
 
 622-8 
 
 623-5 
 
 
 618-5 
 
 619-1 
 
 619-7 
 
 620-3 
 
 620-9 
 
 621-5 
 
 
 616-6 
 
 617-2 
 
 617-8 
 
 618-4 
 
 619-0 
 
 619-6 
 
 
 614-7 
 
 615-3 
 
 615-9 
 
 616-5 
 
 617-1 
 
 617-7 
 
 
 612-8 
 
 613-4 
 
 614-0 
 
 614-6 
 
 615-2 
 
 615-8 
 
 
 610-9 
 
 611-5 
 
 612-2 
 
 612-8 
 
 613-4 
 
 614-0 
 
 
 609-1 
 
 609-7 
 
 610-3 
 
 610-9 
 
 611-5 
 
 612-1 
 
 
 607-2 
 
 607-8 
 
 608-5 
 
 609-1 
 
 609-7 
 
 610-3 
 
 
 605-4 
 
 606-0 
 
 606-6 
 
 607-2 
 
 607-8 
 
 608-4 
 
 
 603-6 
 
 604-2 
 
 604-8 
 
 605-4 
 
 606-0 
 
 606-6 
 
 
 601-8 
 
 602-4 
 
 603-0 
 
 603-6 
 
 604-2 
 
 604-8 
 
 
 600-0 
 
 600-6 
 
 601-2 
 
 601-8 
 
 602-4 
 
 603-0 
 
 
 598-2 
 
 598-8 
 
 599-4 
 
 600-0 
 
 600-6 
 
 601-2 
 
 
 596-4 
 
 597-0 
 
 597-6 
 
 598-2 
 
 598-8 
 
 599-4 
 
 
 594-7 
 
 595-3 
 
 595-9 
 
 596-4 
 
 597-0 
 
 597-6 
 
 
 592-9 
 
 593-5 
 
 594-1 
 
 594-7 
 
 595-3 
 
 595-9 
 
 
 591-2 
 
 591-8 
 
 592-3 
 
 592-9 
 
 593-5 
 
 594-1 
 
 
 589-4 
 
 590-0 
 
 590-6 
 
 591-2 
 
 591-8 
 
 592-4 
 
 
 587-7 
 
 588-3 
 
 588-9 
 
 589-5 
 
 590-1 
 
 590-6 
 
 
 586-0 
 
 586-6 
 
 587-2 
 
 587-8 
 
 588-3 
 
 588-9 
 
 
 584-3 
 
 584-9 
 
 585-5 
 
 586-0 
 
 586-6 
 
 587-2 
 
 
 582-6 
 
 583-2 
 
 583-8 
 
 584-3 
 
 584-9 
 
 585-5 
 
 
 580-9 
 
 581-5 
 
 582-1 
 
 582-7 
 
 583-2 
 
 583-8 
 
 
 579-2 
 
 579-8 
 
 580-4 
 
 581-0 
 
 581-6 
 
 582-1 
 
 
 577-6 
 
 578-2 
 
 578-7 
 
 579-3 
 
 579-9 
 
 580-5 
 
 
 575-9 
 
 576-5 
 
 577-1 
 
 577-6 
 
 578-2 
 
 578-8 
 
 
 574-3 
 
 574-9 
 
 575-4 
 
 576-0 
 
 576-6 
 
 577-1 
 
 
 572-6 
 
 573-2 
 
 573-8 
 
 574-4 
 
 574-9 
 
 575-5 
 
 
 571-0 
 
 571-6 
 
 572-2 
 
 572-7 
 
 573-3 
 
 573-9 
 
 
 569-4 
 
 570-0 
 
 570-5 
 
 571-1 
 
 571-7 
 
 572-2 
 
 
 567-8 
 
 568-4 
 
 568-9 
 
 569-5 
 
 570-1 
 
 570-6 
 
 
 566-2 
 
 566-8 
 
 567-3 
 
 567-9 
 
 568-4 
 
 569-0 
 
 
 564-6 
 
 565-2 
 
 565-7 
 
 566-3 
 
 566-8 
 
 567-4 
 
 
 563-0 
 
 563-6 
 
 564-1 
 
 564-7 
 
 565-3 
 
 565-8 
 
 
 561-4 
 
 562-0 
 
 562-6 
 
 563-1 
 
 563-7 
 
 5t>4-^ 
 
 
 559-9 
 
 560-4 
 
 561-0 
 
 561-6 
 
 562-1 
 
 562-7 
 
 
 558-3 
 
 558-9 
 
 559-4 
 
 560-0 
 
 560-6 
 
 561-1 
 
 
 556-8 
 
 557-3 
 
 557-9 
 
 558-4 
 
 559-0 
 
 559-6 
 
 
 555-2 
 
 555-8 
 
 556-3 
 
 556-9 
 
 557-5 
 
 558-0 
 
 
 553-7 
 
 554-3 
 
 554-8 
 
 555-4 
 
 555-9 
 
 556-5 
 
 
 552-2 
 
 552-7 
 
 553-3 
 
 553-8 
 
 554-4 
 
 554-9 
 
 
 550-7 
 
 551-2 
 
 551-8 
 
 552-3 
 
 552-9 
 
 553-4 
 
 
 549-2 
 
 549-7 
 
 550-3 
 
 550-8 
 
 551-4 
 
 551-9 
 
672 
 
 THE ASSAY OF SILVER. 
 
 COMMON 
 
 Weight of 
 Assay in 
 Milligrs. 
 
 0. 
 
 i. 
 
 O " 
 
 3. 
 
 4. 
 
 
 1835 
 
 545-0 
 
 545-5 
 
 546-0 
 
 546-6 
 
 547-1 
 
 
 1840 
 
 543-5 
 
 544-0 
 
 544-6 
 
 545-1 
 
 545-6 
 
 
 1845 
 
 542-0 
 
 542-5 
 
 543-1 
 
 543-6 
 
 544-2 
 
 
 1850 
 
 540-5 
 
 541-1 
 
 541-6 
 
 542-2 
 
 542-7 
 
 
 1855 
 
 539-1 
 
 539-6 
 
 540-2 
 
 540-7 
 
 541-2 
 
 
 1860 
 
 537-6 
 
 538-2 
 
 538-7 
 
 539-2 
 
 539-8 
 
 
 1865 
 
 536-2 
 
 536-7 
 
 537-3 
 
 537-8 
 
 538-3 
 
 
 1870 
 
 534-8 
 
 535-3 
 
 535-8 
 
 536-4 
 
 536-9 
 
 
 1875 
 
 533-3 
 
 533-9 
 
 534-4 
 
 534-9 
 
 535-5 
 
 
 1880 
 
 531-9 
 
 532-4 
 
 533-0 
 
 533-5 
 
 534-0 
 
 
 1885 
 
 530-5 
 
 531-0 
 
 531-6 
 
 532-1 
 
 532-6 
 
 
 1890 
 
 529-1 
 
 529-6 
 
 530-2 
 
 530-7 
 
 531-2 
 
 
 1895 
 
 527-7 
 
 528-2 
 
 528-8 
 
 529-3 
 
 529-8 
 
 
 1900 
 
 526-3 
 
 526-8 
 
 527-4 
 
 527-9 
 
 528-4 
 
 
 1905 
 
 524-9 
 
 525-4 
 
 526-0 
 
 526-5 
 
 527-0 
 
 
 1910 
 
 523-6 
 
 524-1 
 
 524-6 
 
 525-1 
 
 525-6 
 
 
 1915 
 
 522-2 
 
 522-7 
 
 523-2 
 
 523-8 
 
 524-3 
 
 
 1920 
 
 520-8 
 
 521-3 
 
 521-9 
 
 522-4 
 
 522-9 
 
 
 1925 
 
 519-5 
 
 520-0 
 
 520-5 
 
 521-0 
 
 521-6 
 
 
 1930 
 
 518-1 
 
 518-6 
 
 519-2 
 
 519-7 
 
 520-2 
 
 
 1935 
 
 516-8 
 
 517-3 
 
 517-8 
 
 518-3 
 
 518-9 
 
 
 1940 
 
 515-5 
 
 516-0 
 
 516-5 
 
 517-0 
 
 517-5 
 
 
 1945 
 
 514-1 
 
 514-6 
 
 515-2 
 
 515-7 
 
 516-2 
 
 
 1950 
 
 512-8 
 
 513-3 
 
 513-8 
 
 514-4 
 
 514-9 
 
 
 1955 
 
 511-5 
 
 512-0 
 
 512-5 
 
 513-0 
 
 513-5 
 
 
 1960 
 
 510-2 
 
 510-7 
 
 511-2 
 
 511-7 
 
 512-2 
 
 
 1965 
 
 508-9 
 
 509-4 
 
 509-9 
 
 510-4 
 
 510-9 
 
 
 1970 
 
 507-6 
 
 508-1 
 
 508-6 
 
 509-1 
 
 509-6 
 
 
 1975 
 
 506-3 
 
 506-8 
 
 507-3 
 
 507-8 
 
 508-3 
 
 
 1980 
 
 505-0 
 
 505-6 
 
 506-1 
 
 506-6 
 
 507-1 
 
 
 1985 
 
 503-8 
 
 504-3 
 
 504-8 
 
 505-3 
 
 505-8 
 
 
 1990 
 
 502-5 
 
 503-0 
 
 503-5 
 
 504-0 
 
 504-5 
 
 
 1995 
 
 501-3 
 
 501-8 
 
 502-3 
 
 502-8 
 
 503-3 
 
 
 2000 
 
 500-0 
 
 500-5 
 
 501-0 
 
 501-5 
 
 502-0 
 
 
TABLE FOR THE WET ASSAY OF SILVER. 
 
 673 
 
 SALT continued. 
 
 5. 
 
 G. 
 
 I . 
 
 8. 
 
 9. 
 
 10. 
 
 547-7 
 
 548-2 
 
 548-8 
 
 549-3 
 
 549-9 
 
 550-4 
 
 546-2 
 
 546-7 
 
 547-3 
 
 547-8 
 
 548-4 
 
 548-9 
 
 544-7 
 
 545-3 
 
 545-8 
 
 546-3 
 
 546-9 
 
 547-4 
 
 543-2 
 
 543-8 
 
 544-3 
 
 544-9 
 
 545-4 
 
 545-9 
 
 541-8 
 
 542-3 
 
 542-9 
 
 543-4 
 
 543-9 
 
 544-5 
 
 540-3 
 
 540-9 
 
 541-4 
 
 541-9 
 
 542-5 
 
 543-0 
 
 538-9 
 
 539-4 
 
 539-9 
 
 540-5 
 
 541-0 
 
 541-5 
 
 537-4 
 
 538-0 
 
 538-5 
 
 539-0 
 
 539-6 
 
 540-1 
 
 536-0 
 
 536-5 
 
 537-1 
 
 537-6 
 
 538-1 
 
 538-7 
 
 534-6 
 
 535-1 
 
 535-6 
 
 536-2 
 
 536-7 
 
 537-2 
 
 533-2 
 
 533-7 
 
 534-2 
 
 534-7 
 
 535-3 
 
 535-8 
 
 531-7 
 
 532-3 
 
 532-8 
 
 533-3 
 
 533-9 
 
 534-4 
 
 530-3 
 
 530-9 
 
 531-4 
 
 531-9 
 
 532-4 
 
 533-0 
 
 528-9 
 
 529-5 
 
 530-0 
 
 530-5 
 
 531-0 
 
 531-6 
 
 527-6 
 
 528-1 
 
 528-6 
 
 529-1 
 
 529-7 
 
 530-2 
 
 526-2 
 
 526-7 
 
 527-2 
 
 527-7 
 
 528-3 
 
 528-8 
 
 524-8 
 
 525-3 
 
 525-8 
 
 526-4 
 
 526-9 
 
 527-4 
 
 523-4 
 
 524-0 
 
 524-5 
 
 525-0 
 
 525-5 
 
 526-0 
 
 522-1 
 
 522-6 
 
 523-1 
 
 523-6 
 
 524-2 
 
 524-7 
 
 520-7 
 
 521-2 
 
 521-8 
 
 522-3 
 
 522-8 
 
 523-3 
 
 519-4 
 
 519-9 
 
 520-4 
 
 520-9 
 
 521-4 
 
 522-0 
 
 518-0 
 
 518-6 
 
 519-1 
 
 519-6 
 
 520-1 
 
 520-6 
 
 516-7 
 
 517-2 
 
 517-7 
 
 518-2 
 
 518-8 
 
 519-3 
 
 515-4 
 
 515-9 
 
 516-4 
 
 516-9 
 
 517-4 
 
 517-9 
 
 514-1 
 
 514-6 
 
 515-1 
 
 515-6 
 
 516-1 
 
 516-6 
 
 512-8 
 
 513-3 
 
 513-8 
 
 514-3 
 
 514-8 
 
 515-3 
 
 511-4 
 
 512-0 
 
 512-5 
 
 513-0 
 
 513-5 
 
 514-0 
 
 510-1 
 
 510-7 
 
 511-2 
 
 511-7 
 
 512-2 
 
 512-7 
 
 508-9 
 
 509-4 
 
 509-9 
 
 510-4 
 
 510-9 
 
 511-4 
 
 507-6 
 
 508-1 
 
 508-6 
 
 509-1 
 
 509-6 
 
 510-1 
 
 506-3 
 
 506-8 
 
 507-3 
 
 507-8 
 
 508-3 
 
 508-8 
 
 505-0 
 
 505-5 
 
 506-0 
 
 506-5 
 
 507-0 
 
 507-5 
 
 503-8 
 
 504-3 
 
 504-8 
 
 505-3 
 
 505-8 
 
 506-3 
 
 502-5 
 
 503-0 
 
 503-5 
 
 504-0 
 
 504-5 
 
 505-0 
 
 XX 
 
674 THE ASSAY OF SILVER. 
 
 APPLICATION. 
 
 Assay of Pure, or nearly Pure, Silver, the Temperature of 
 the Normal Solution of Salt being that at which it was 
 standardised. 
 
 First example. Let the ingot of silver have an ap- 
 proximative standard of from 995 to 1000 thousandths. 
 Take one gramme ; dissolve it in ten grammes of nitric 
 acid, in the bottle, fig. 118. Then pour into the bottle an 
 exact measure of the normal solution of salt, and brighten 
 by agitation. The silver not being supposed to be quite 
 pure, the standard is not further sought for by the decline 
 solution of salt, but that of silver nitrate is employed. 
 
 One thousandth of this latter solution is poured into 
 the bottle ; it becomes cloudy, and is well agitated. A 
 second and a third thousandth also give a precipitate, but 
 not so a fourth. From these data the following is the 
 method of ascertaining the standard of the alloy : 
 
 The last thousandth of the decime solution of silver, 
 having produced no cloudiness, is not to be counted. The 
 third was necessary, but only partially so ; consequently 
 the number of thousandths of silver necessary to decom- 
 pose the excess of salt is more than 2 and less than 3 ; in 
 other words, it is equal to the mean, 21 ; but since 2-| 
 thousandths of silver have been required to complete the 
 precipitation of salt representing 1000 thousandths of 
 silver, it is evident that the silver submitted to assay con- 
 tained 21 thousandths of alloy, and that its standard, to 
 within nearly half a thousandth, is but 997-J. 
 
 If it be considered necessary to arrive nearer the true 
 standard, the following proofs must be employed : Pour 
 into the solution 11 thousandths of salt, which will decom- 
 pose a like number of thousandths of silver.* After due 
 
 It has already been stated how a thousandth of the decime solution may 
 be subdivided by the number of drops furnished by the pipette. If, for in- 
 stance, it contains 20 drops, 10 will give the half, 5 the quarter, &c. Half a 
 thousandth may also be obtained by diluting the solution with its volume of 
 water, and using a whole pipetteful. This latter plan has been found the best 
 in practice. 
 
ASSAY OF PURE, OR NEARLY PURE, SILVER. 675 
 
 .agitation, add half a thousandth of silver nitrate. Sup- 
 posing a cloudiness is produced, no further addition must 
 be made ; for it is already known that above the third 
 thousandth no precipitate is formed in the liquid by silver 
 nitrate, and consequently only half of the last half-thou- 
 sandth must be calculated, as only a portion of it was 
 necessary. From which, the entire number of thousandths 
 of silver nitrate being 4J, and those of salt 1-J, there 
 remains 2f for the number of thousandths of nitrate of 
 silver addded to the normalsolution ; and consequently the 
 standard of the aUoy is 1000 -2f =997J. If, on the other 
 hand, the last half-thousandth of the silver nitrate had 
 produced no cloudiness, it would not have to be reckoned, 
 and only half of the preceding half-thousandth would have 
 been taken. Thus from the 4 thousandths of silver 
 nitrate employed a quarter of a thousandth is deducted ; 
 and from the difference, of, is yet deducted 1-J of salt, the 
 final remainder being 2^ thousandths of silver nitrate which 
 have been added to the normal solution : the standard of 
 the alloy would be 1000 -2J= 997 j. 
 
 Although the above-described operation is very simple, 
 yet it is desirable, in order to avoid al] confusion, to note 
 in writing such thousandths of salt or silver nitrate as are 
 added. The thousandths of salt indicating an increase of 
 standard should be preceded by the sign + ; and the thou- 
 sandths of silver nitrate announcing a diminution of stan- 
 dard, by the sign . 
 
 Second Example. Suppose the ingot has a presumed 
 standard of 895 thousandths, and the temperature of the 
 normal solution supposed invariable. 
 
 Find in the table of standards (Salt Table), first column, 
 that which approaches the nearest to 895 ; it will be 
 found to be 896- 9, corresponding to the weight of 1115 
 milligrammes. This weight of the alloy is taken and 
 dissolved in nitric acid, a measure of normal solution of 
 salt added, and the whole well agitated. The operator is, 
 however, doubtful whether the assay must be proceeded 
 with by the decime salt solution, or the silver nitrate 
 decime solution. If the former produces a precipitate^ 
 
 x x2 
 
676 THE ASSAY OF SILVER. 
 
 it is gone on with ; but if it does not precipitate, that 
 already added is decomposed by a similar addition of the 
 second, and the solution rendered bright by agitation. A 
 starting-point has now been arrived at for the continuance 
 of the assay, for it is known that the silver nitrate solution 
 must be employed. 
 
 Suppose, then, that the alloy, after the addition of the 
 measure of normal solution, yet gives a precipitate with 
 the decime solution of salt. The first 5 thousandths 
 produce a precipitate, but not the sixth, which conse- 
 quently is not counted. The fifth has only been partially 
 required, so that it is more than 4 thousandths, and less 
 than 5, or the mean, 4^, is the quantity required to 
 entirely precipitate the excess of silver in the alloy sub- 
 mitted to assay. But by neglecting at first the fraction 
 0*5, seek in the Salt Table of Standards the number found 
 on the longitudinal line of the weight 1115, under column 
 4 ; it is 900-4, and on adding 0-5 to this number we have 
 900-9, or 901, for the required standard. 
 
 Supposing, however, that the same alloy, after the 
 addition of the normal measure of salt, gives a precipitate 
 with silver nitrate, and that the 3 first thousandths pro- 
 duce a cloudiness, but not the fourth. The number of 
 thousandths of silver nitrate really necessary for complete 
 precipitation will be very nearly 2-J. To ascertain the 
 real standard of the alloy of which 1115 thousandths were 
 supposed to contain about 1000 thousandths of silver, take 
 the number found in the horizontal line 1115, and in the 
 column 2 of the Silver Nitrate Table. This number, which 
 is 895-1, diminished by the fraction 0-5, gives 894*6 for 
 the standard of the alloy to within half a thousandth. 
 
 Third Example. The actual temperature of the normal 
 solution of salt being 18, when it was standardised at 15. 
 
 The ingot of silver submitted to assay has an approxi- 
 mative standard of 795 thousandths. Find in the Salt 
 Table of Standards, first column, that which is nearest 
 to it ; it is 793*7, corresponding to the weight 1260. 
 This weight of the alloy is taken, and the operation pro- 
 ceeded with as already described. Supposing it had 
 
GRADUATION OF THE NORMAL SOLUTION OF SALT. 077 
 
 required 6'5 thousandths of salt to precipitate the whole 
 of the silver contained in the alloy to within half a 
 thousandth, the required standard, without correction 
 for temperature, will be 798-4 + 0'4 = 798'8. But, making 
 this correction, recourse must be had to the table, page 
 651, column 15 : the number 0'3, which will be found in 
 the horizontal line 18 and the column 15, possesses the 
 sign ; consequently it must be deducted from 798*8, and 
 the remainder, 798'5, will be the standard weight. If the 
 temperature of the solution, instead of being 3 higher than 
 at the time it was standardised, was 3 lower, that is 12, 
 the correction must be added, and would be equal to +0-2. 
 The standard of the alloy would conse- FrG 12 4. 
 quently be 798-8 + 0-2 = 799. 
 
 Graduation of the Normal Solution of Salt, 
 the Temperature being different to that at 
 which it was wished to be graduated. 
 
 Two equally ready processes can be 
 employed. The one consists in reducing 
 the temperature of the solution to the 
 desired degree before standardising ; the 
 other, in estimating its standard without 
 regard to the temperature of the solution, 
 and then correcting its influence by the 
 aid of the tables of correction already 
 given. 
 
 First Process. Place the liquid to be 
 graduated in a bottle, F, fig. 124. Intro- 
 duce a thermometer, and heat to a deter- 
 minate degree, say 20 for instance. This 
 done, place the jet of the pipette in the 
 bottle ; raise the liquid by aspiration by 
 means of the conical tube, T, fig. 124, 
 which is adapted to the opening of the 
 air-cock, R. As soon as the liquid is 
 raised a little, above the mark a 6, which determines the 
 capacity of the pipette, close the stopcock, and complete 
 
678 THE ASSAY OF SILVEK. 
 
 the measurement as usual. This same means of filling 
 the pipette by aspiration may be employed to fill it either 
 with caustic alkali or nitric acid, as the case may be, to 
 cleanse it instead of taking it to pieces. 
 
 Second Method. The solution of salt being supposed 
 at a temperature of 16, and it is desired to graduate it 
 at that of 20. Proceed with the standardising without 
 regard to temperature ; but when it is obtained in each 
 trial assay, it is necessary to make the correction required 
 by the temperature. 
 
 If, for example, in an approximative assay the standard 
 of the solution was expressed by 1001*5 this standard 
 would not only be too weak by 1*5 thousandth, but, 
 according to the table of temperatures, by yet another 
 0-5, for the solution is weakened by this quantity, by 
 passing from 16 to 20. The standard, if taken at this 
 last temperature, would be too low by 2 thousandths, and 
 must consequently be corrected. 
 
 If, on the other hand, the standard of the solution were 
 too high instead of too low, and expressed by 998*5 at 
 the temperature of 16 ; at that of 20, the solution being 
 weakened by 0'5, the standard would only be but one 
 thousandth too high, and it must be corrected by that 
 quantity. 
 
 Approximative Estimation of the Standard of an 
 Unknown Alloy. 
 
 It has always been supposed, in the experiments already 
 detailed, that the approximative standard of the alloy sub- 
 mitted to assay was known : and this, indeed, is nearly 
 always the case. If, however, this be unknown, two 
 means are available for obtaining the necessary know- 
 ledge. A decigramme of the alloy is cupelled with one 
 gramme of lead ; or if it be desirable not to use the cupel, 
 it may be ascertained by the wet method, in the follow- 
 ing manner : 
 
 The assayer supposes the standard of the alloy known 
 to be about a twentieth, and it can always be found nearer 
 than that by touch, density, &c. A weight relative to its 
 
MODES OF ABRIDGING MANIPULATION. 679 
 
 supposed standard is taken, and its standard sought by 
 adding the decime liquid by 10 thousandths at a time, 
 by means of pipettes of this capacity (see fig. Fm 125 
 125). The term of complete precipitation is 
 soon passed, and the standard of the alloy to 
 about 5 thousandths is thus ascertained. The 
 approximate standard to 2-J- thousandths may 
 be obtained by adding only 5 thousandths oi 
 solution at a time. 
 
 Suppose the alloy 840 thousandths. . Take 
 the weight 1190, corresponding to this standard, 
 and proceed "as in an ordinary assay, adding 
 each time, for example, a pipette of 10 thou- 
 sandths of salt solution. It is found the fifth 
 pipette gives no precipitate, and consequently 
 the number of thousandths of salt for the pre- 
 cipitate of the silver to within 5 thousandths is 
 35. The 1199 of alloy will therefore contain 
 1000 + 35 = 1035 of silver ; and the approxi- 
 mative standard will be obtained by the pro- 
 portion 
 
 1120: 1035:: 1000:0=869-7. 
 
 Modes of Abridging Manipulation. 
 
 In the statement already given of the mode of con- 
 ducting the assay by the wet method, only such instructions 
 have been given as were necessary for its full comprehen- 
 sion, and everything that might call away or FIG. 126. 
 fatigue the attention has been omitted. Never- 
 theless, here it will be convenient to describe 
 some methods of abridging the necessary mani- 
 pulations, supposing that ten, or at least five, 
 assays are made at once. 
 
 Bottles. It is necessary to have these all, 
 as nearly as possible, of the same height and 
 diameter. They are marked progressively on the shoulder, 
 as are also their stoppers (fig. 126), thus 1, 2, 3, 4, &c. 
 They are taken successively by tens, in the natural order. 
 
680 
 
 THE ASSAY OF SILVER. 
 
 The stoppers are placed on a support, numbered in the same 
 manner (fig. 127). The support is pierced with ten holes, 
 127. distinguished in precedence by 
 
 a mark between the fifth and 
 
 sixth. 
 
 Stand. Each ten flasks 
 
 are in turn placed in a case or 
 stand of japanned tin-plate (fig. 128), having ten compart- 
 ments numbered from 1 to 10. Each of these compart- 
 ments is cut anteriorly to about half its length, so as to 
 
 FIG. 128. 
 
 FIG. 129. 
 
 FIG. 130. 
 
 allow the numbers, of the bottles to be seen. The same 
 stand serves for all the series, by making the same units 
 correspond : thus, No. 23 of the third series is placed in 
 
 stand No. 3, &c. When each 
 flask is charged with the 
 alloy, about 10 grammes of 
 nitric acid, 40 C., are mea- 
 sured by a pipette (fig. 125) 
 introduced into the bottles by 
 means of a funnel with a large 
 neck (fig. 129). The whole 
 are then exposed to the heat 
 of a water-bath, to facilitate 
 the solution of the alloy. 
 
 Water-bath. This is an 
 oblong tin-plate vessel, cal- 
 culated to receive 10 bottles (fig. 130). It has a move- 
 able double bottom, pierced with small holes,, the prin- 
 
MODES OF ABE1DGING MANIPULATION. 681 
 
 cipal object of which is to prevent the fracture of the 
 bottles by isolating them from the bottom of the vessel, 
 which is immediately exposed to the heat. On the 
 moveable bottom are soldered the cylinders c c, three or 
 four centimetres in height, and above which, at the distance 
 of eight centimetres, is a sheet of tin plate, p p, pierced with 
 ten holes, corresponding to the cylinders, and connected 
 with the movable bottom by the supports, s s. These 
 cylinders, and the sheet of tin plate, are destined to isolate 
 the bottles, F F, one from the other in the bath, and to 
 keep them some time suspended over it, when the water is 
 boiling, before complete immersion. The water-bath may 
 be replaced by a steam-bath ; the bottles will then be sup- 
 ported by a grating above the surface of the water. The 
 solution of the alloy in the nitric acid takes place rapidly, 
 and as it gives rise to an abundant evolution of nitrous 
 vapour, it must be made under a flue having a good 
 draught. 
 
 Flue. This is represented at fig. 131. C C is a flue 
 resting on a table or support, T T, about 90 centimetres 
 high. The anterior side in the FlG 1B1 
 
 figure is removed to show^ the 
 water-bath 5, and the furnace 
 F. The opening, 0, of the flue 
 is closed by the wooden door, p, 
 movable on two eccentric pivots, 
 which keep it up during the 
 solution, and allow it to fall so 
 that the flask may be placed 
 upon it. The nitrous vapour is 
 removed from the bottles with the blower (fig. 123). The 
 hood, ff, prevents the diffusion of the nitrous vapour in 
 the laboratory.. 
 
 Agitator. Figure 132 gives a sufficiently exact idea of 
 this apparatus, and dispenses with a long description. It 
 has ten cylindrical compartments, numbering from 1 to 10. 
 The bottles, after solution of the alloy, are placed in it in 
 the order of their numbers. The agitator is then placed 
 by the side of the pipette, by which is measured the normal 
 
682 
 
 THE ASSAY OF SILVER. 
 
 solution ol salt, and into each flask is poured a pipetteful 
 of the solution. The bottles are fitted with their stoppers, 
 previously moistened with distilled water (fig. 133); they 
 are then fixed in order with wooden wedges (fig. 134). 
 The agitator is suspended to a spring, 7?, and a rapid 
 alternating movement given to it with both hands, by which 
 the solution is agitated, and in less than a minute rendered 
 
 FIG. 132. 
 
 FIG. 133. 
 
 FIG. 134. 
 
 as clear as water. The movement is assisted by a spiral 
 spring, B, fixed to the agitator and its stand. The agitation 
 finished, the wedges are removed, and placed in the vacant 
 spaces between the compartments. The agitator is taken 
 from the spring, and the bottles^placed in order on a table 
 prepared to receive them. 
 
 Table. This table (fig. 135) has a double bottom ; the 
 upper is pierced with ten holes, a little larger than the 
 
MODES OF ABKIDGING MANIPULATION. 
 
 083- 
 
 diameter of the bottles, and of such a distance from the 
 lower portion, or false bottom, that the flasks do not rise 
 above its edge, or at least but little. This disposition is to 
 
 FIG. 135. 
 
 FIG. 136. 
 
 protect the silver chloride from the light, for it decomposes 
 in contact with water, and a little hydrochloric acid is 
 produced, which requires for its precipitation a certain 
 quantity of silver nitrate, and so lowers the standard of 
 the alloy. This cause of error is, however, not very great, 
 at least when the light does not fall directly on the 
 chloride ; but it is easy to avoid, and should not be neg- 
 lected. The disposition already pointed out does not at all 
 complicate the process, and is moreover useful, as it pre- 
 vents the fracture or upsetting of the bottles. When but 
 one bottle is operated on, it is placed for agitation in a 
 japanned tin-plate cylinder, which is held as shown at fig. 
 136. On placing the bottles in their respective places on 
 the table, a brisk circular movement is given to them, so as 
 to remove any silver chloride 
 adhering to the sides ; their 
 stoppers are removed and sus- 
 pended by spring pincers, a a. 
 These are formed of sheet-iron 
 wire (see fig. 137). A thou- 
 sandth of the decime solution is then poured into each 
 bottle, and before this has been completed there will have 
 
 FIG. 137. 
 
<>84 THE ASSAY OF SILVER. 
 
 formed in the first bottles where there is any precipitate, a 
 well-marked nebular layer about a centimetre in thickness. 
 At the back of the table is a black board, PP, divided 
 into compartments numbered from 1 to 10, on each of 
 which is marked with chalk the number of thousandths of 
 decime liquid added to the contents of the corresponding 
 bottle. The thousandths of salt announcing augmentation 
 of standard are preceded by the sign + , those of silver 
 nitrate by the sign . 
 
 Lastly, the black board carries a small shelf pierced 
 with holes, 1 1, and these receive the funnels or drain the 
 bottles ; on this shelf also are fastened the pincers for sus- 
 taining the stoppers. 
 
 Cleaning the Bottles. The assays terminated, the liquid 
 from each flask is poured into a large vessel in which there 
 is always a slight excess of common salt ; and when it is 
 full, the clear supernatant fluid is removed by means of a 
 syphon. (Immediately will be given the means of reducing 
 the silver chloride so collected to the metallic state.) The 
 bottles, to the number of ten, are first rinsed with the same 
 water passed from one to the other, then a second, and then 
 a third time with fresh water. They are then placed to 
 FIG 138 drain on the board just mentioned, 
 
 and the stoppers are placed in a 
 stand by series of tens (see figs. 138 
 and 127). It is important to remark, 
 that when a glass has been rinsed 
 with distilled water, care must be 
 taken not to rub it with the fingers, for water poured in 
 such a vessel would always be clouded on the addition of 
 silver nitrate. This effect is due to the chlorides contained 
 in the perspiration, and is of course more to be guarded 
 against in summer. 
 
 Reduction of Silver Chloride , obtained in the Assay of 
 Alloys by the Wet Method. 
 
 Silver chloride can be reduced without sensible loss, 
 after haying been well washed, by plunging it into scraps 
 
PREPARATION OF PUKE SILVER. 685 
 
 of iron or zinc, and adding dilute sulphuric acid in suffi- 
 cient quantity to set up a slight disengagement .of hydro- 
 gen gas. The whole can be left to itself, and in the course 
 of a few days the silver is completely reduced. This point 
 can be easily estimated by the colour and nature of the 
 product, but better still by treating a small quantity with 
 ammonia, which, if the chloride is perfectly reduced, will 
 give no precipitate or cloudiness on treatment with an acid. 
 The chlorine remains in solution in the water combined 
 with zinc or iron. The residue must now be washed ; the 
 first washings are made with acidulated water, to dissolve 
 iron oxide which might have formed, and the following 
 with ordinary water : after having completed the washing 
 as much water as may be left is decanted, the mass dried, 
 and a little powdered borax added. Nothing now remains 
 but to fuse it. The powdered silver being voluminous, it 
 is placed by separate portions in the crucible, in propor- 
 tion as it sinks. The heat should at first be moderate, but 
 towards the end of the operation should be sufficiently 
 high to reduce the silver and slag to a state of complete 
 liquidity. If it be found that not quite all the chloride 
 was reduced by the iron or zinc, a little potassium or 
 sodium carbonate may be added to the powdered silver. 
 The standard of silver thus obtained is from 999 to 1000 
 thousandths. 
 
 Preparation of Pure Silver. 
 
 Take the silver prepared as above, dissolve it in nitric 
 acid, and leave the solution some time in perfect rest in 
 water, to deposit any gold it might contain. Decant the 
 solution, and precipitate with common salt, well wash the 
 precipitate, and reduce it, when the resulting silver will be 
 pure. 
 
 M. Gay-Lussac here gives a description of a process for 
 the precipitation of chlorine from nitric acid for use in the 
 mode of assay already described ; but as that acid in a state 
 of ordinary purity forms an article of commerce, and can be 
 obtained at most operative chemists, the process will riot be 
 here reproduced. 
 
C86 THE ASSAY OF SILVER. 
 
 Modifications required in the Assay of Silver Alloys 
 containing Mercury. 
 
 Whenever mercury is present in solution with silver, it 
 is thrown down as insoluble chloride, and the assay ren- 
 dered inaccurate. The presence of mercury in silver can 
 be readily detected by the remarkable change which occurs 
 in silver chloride on exposure to light (viz. blackening) 
 when free from mercury ; but if the smallest quantity of 
 the latter metal be present, no blackening will ensue. This 
 source of error was removed by M. Levol in the following 
 manner : The sample being dissolved, as usual, in nitric 
 acid, it was supersaturated with 25 cubic centimetres of 
 caustic ammonia ; the pipetteful of normal solution was 
 then added, and the excess of ammonia supersaturated 
 with 20 cubic centimetres of acetic acid, and the operation 
 continued in the usual way. 
 
 Some little time after the publication of this, M. Gray- 
 Lussac examined the above process himself, and very con- 
 siderably simplified it. He says : ' After having confirmed 
 by several experiments the accuracy of M. Level's pro- 
 cess, I thought it might be simplified by adding to the 
 nitric solution of silver the ammonia and acetic acid at one 
 and the same time, but in sufficient quantity to saturate 
 the whole of the nitric acid, both that in combination with 
 the silver and that in the free state. Ten grammes of ammo- 
 nium acetate, were added, with a little water, to the silver 
 dissolved in nitric acid, and the assay finished in the 
 ordinary manner. The quantity indicated by synthesis 
 was found very accurately, although 100 thousandths of 
 mercury had been added/ Finally, M. Gay-Lussac found 
 that 10 grammes of sodium acetate, in crystals, also fully 
 answered the purpose ; and as that is a very cheap com- 
 mercial salt, it is the best adapted for overcoming the diffi- 
 culty in this class of assay, as regards the presence of 
 mercury. 
 
METHOD OF ASSAYING SILVER BARS. 
 
 087 
 
 FIG. 139 
 
 Method of Taking the Assay from the Ingot. 
 
 The ingots are so rarely perfectly homogeneous, even 
 taking as a starting-point the standard 950 thousandths, 
 that the differences remarked between the assays of samples 
 made in different places should 
 rather be attributed to the 
 above cause than to the assay 
 itself. It is important, there- 
 fore, to take a sample in a 
 uniform manner, and from 
 the same depth, on the upper 
 surface of the ingot as on the 
 lower. This condition is per- 
 fectly fulfilled by boring the 
 ingots with a kind of drill, 
 similar to that employed by 
 the smith, and which is re- 
 presented at fig. 139. The 
 ingot, Z, is placed in a copper 
 tray, C\ and in order to 
 retain the borings, which 
 might otherwise be thrown 
 out, the drill,/, is surrounded 
 by a casing, m, which does 
 not impede its motion, and 
 stands freely on the ingot. 
 After a few turns of the drill, 
 
 the first borings, which are not pure, are removed by 
 means of a feather, and only those following are collected 
 and reserved for assay. If it be desirable to try the lateral 
 faces, it is necessary to employ a pressure screw, to keep 
 the ingot in the position that may be deemed necessary. 
 
 Method of Assaying Silver Bars adopted in the Assay 
 Offices of II.M. Indian Mints. 
 
 According to the report of H. E. Busteed, M.D., H.M. 
 Madras Army, refining assay master, Calcutta Mint, the 
 
038 THE ASSAY OF SILVER. 
 
 silver bars are assayed by estimating the silver as chlo- 
 ride. The cupellation assay is not correct enough to 
 satisfy the sellers and purchasers : the Gay-Lussac assay 
 is not used, because most of the bullion contains mercury, 
 lead, and other base metals ; because a previous cupel- 
 lation assay is required ; because the high temperature of 
 the climate Causes an evaporation of the salt solution ; and 
 because a large number of persons would be necessary on 
 account of the large daily number of assays. 
 
 The credit is due to Mr. J. Dodd, a former assay 
 master of the Calcutta Mint (and a surgeon in the 
 Madras army), of having overcome those difficulties of 
 manipulation, inasmuch as he modified and simplified 
 them, and in short so systematised the whole practical 
 working of the process as to render its application to 
 the assaying of silver, to any amount, easy, accurate, 
 and economical. 
 
 The process is technically known as the ' chloride 
 process.' It is thus described by Dr. Busteed in the 
 ' Chemical News ' for November 1871 : 
 
 The samples (or ' musters ') for assay are, to save time, 
 first approximately weighed by an assistant ; they are then 
 placed (each sample in duplicate) in small shallow saucers 
 of polished copper, and so brought in batches of 40 on a 
 board, containing in numerical order receptacles for the 
 little saucer, to the assay master, who, in the delicate assay 
 balance, exactly brings each sample to the one required 
 weight.* 
 
 As each sample is weighed, it is transferred from the 
 platinum skiff of the balance to a bottle on the left hand 
 of the assayer, by means of a small copper funnel. The 
 bottles f for this purpose are held in readiness for the 
 musters by an assistant, and, on receiving them, are 
 removed into the laboratory in batches of six. 
 
 On being taken to the laboratory, they are ranged on 
 a circular platform or turn-table, and there one of the 
 
 * The amount of this weight will be more particularly referred to farther on. 
 t The chief appliances will be described more fully at the conclusion of 
 the description of the process. 
 
THE CHLOKIDE PROCESS. 689 
 
 (European) assistants adds by means of a pipette 1^ drm. 
 of nitric acid to each bottle, which are then (without their 
 stoppers) transferred to a sand-bath and exposed to a 
 considerable degree of heat, till solution of the contents is 
 effected. 
 
 The specific gravity of the nitric acid used is generally 
 1-200, i.e. in the case of known alloys of only copper and 
 silver, such as the standard meltings, coins, &c. ; but 
 when the nature of the alloy is uncertain, such as bazaar 
 silver, or some sycee (where the presence of mercury may 
 be suspected), a stronger acid of sp. gr. 1*320 is used. It 
 has been found, too, by experience, that the chlorides 
 from fine bar silver agglomerate better when the solution 
 has been effected in the stronger acid. 
 
 When the samples have been completely dissolved,* 
 the bottles are brought back to the platform, and there 
 each receives through a glass funnel f about six ounces of 
 cold distilled water. 
 
 There is then added to each bottle, through a glass 
 pipette as before, 1^ drm. of hydrochloric acid, sp. gr. 
 TOGO, which immediately converts the silver present into 
 the characteristic white precipitate of silver chloride, which 
 forms in slow-falling curdy volumes. 
 
 The stoppers (previously dipped in distilled water) are 
 then carefully replaced, and the bottles are allowed to 
 stand for five minutes. 
 
 The bottles are next well shaken two and two by the 
 laboratory workman for three or four minutes, till the 
 chloride aggregates and rapidly falls down ; any particles 
 which may remain attached to the neck or upper part of 
 the bottles are washed down by a quick circular motion, 
 and, more distilled water being added to within about 
 two inches of the neck (great caution being observed in 
 removing and returning the stoppers), the bottles are then 
 allowed to rest each in its assigned place on the platform 
 for four hours. 
 
 * A slight residuum of gold, as a black powder, is very generally seen. 
 
 t The portion of this which enters the neck of the bottle is protected or 
 sheathed, with an inch of india-rubber tubing, to prevent chipping, if struck 
 against the neck of the bottle. 
 
 Y Y 
 
690 THE ASSAY OF SILVER. 
 
 At the expiration of that period, the clear supernatant 
 liquid (blue-coloured when copper is present) is removed 
 by a glass syphon, which is lowered to within an inch of 
 the deposited chloride, the greatest care being taken that 
 none of it is drawn up into the syphon. As each platform 
 is made to revolve on its centre, according as each bottle 
 is syphoned, the operator sitting in one place brings the 
 platform round till the next bottle in order gets under the 
 syphon, which is thus in turn lowered into each. The 
 fluid escapes from the long leg of the syphon through a 
 funnel fitted in the table to ajar placed underneath. 
 
 After the first syphoning, the bottles are immediately 
 filled again with distilled water, and each gets a quiet cir- 
 cular motion for a few moments, and the precipitate is again 
 allowed to settle as evenly as possible ; this time it will be 
 sufficient to allow them to rest for two hours, when they 
 are again syphoned as before and the stoppers returned. 
 
 Under ordinary circumstances these two washings are 
 sufficient ; but if the silver is evidently ' coarse,' a third or 
 fourth washing is similarly given. 
 
 When it is considered that the chlorides have been 
 sufficiently washed, the bottles are placed for half an hour 
 in a reclining position on their platforms ; this causes the 
 chloride to fall and settle to one spot, and renders its 
 removal from the bottles more easy. 
 
 Meantime a pneumatic trough has been got ready, 
 capable of containing a batch of twenty inverted bottles ; 
 the trough is filled with distilled water ; for each bottle 
 there is placed on the floor of the trough a small porcelain 
 saucer holding a little Wedgwood crucible or cup, each 
 numbered to correspond to the bottles. A laboratory 
 workman then removes the stoppers from the bottles, and 
 hands them one by one to an assistant at the trough, who, 
 placing his forefinger over the mouth of each bottle, 
 inverts it over its corresponding cup, and does not remove 
 his finger till the neck of the bottle has passed down 
 through the water and well into the cup ; then the finger 
 being taken away, the bulk of the chloride falls by its 
 own weight to the bottom of the cup. 
 
THE CHLORIDE PROCESS. 691 
 
 The bottle is held in the position by two rings, one 
 (the larger) above the other, which are fixed to the sides 
 of the trough ; this arrangement retains each bottle in 
 situ, at the proper slant, and admits of the operator gently 
 revolving or slightly raising the bottle with his left hand, 
 while with the right he patiently taps the bottom and 
 sides till the whole of the chloride has been dexterously 
 got out ; the finger is then again placed over the mouth, 
 and the bottle raised up through the rings and handed 
 (mouth upwards) to the assayer, or to the supervising 
 assistant standing by, who carefully examines it to see 
 that every particle of chloride has been dropped into the 
 cup. When this part of the manipulation has been neatly 
 done, none of the chloride falls over into the saucer which 
 is placed as a precautionary measure under each cup. 
 
 When the chloride falls into the cup, it is in an uneven 
 lumpy state, and not in a favourable condition for being 
 uniformly dried ; it has, therefore, next to be broken up. 
 For this purpose the cups (containing the chlorides, and 
 water to the brim) on removal from the trough are taken 
 in batches on a tray to an assistant seated at a steady 
 table, who first carefully decants off about half the water, 
 and then with a finely polished glass rod (four inches long 
 and one-third of an inch thick) gently stirs and beats the 
 lumpy precipitate, while revolving the cup on the table ; 
 this causes it to lie evenly and loosely at the bottom of 
 the cup as a purplish grey powder, not too fine. 
 
 He next washes the rod over the cup with distilled 
 water from a drop bottle, lest any of the chloride may be 
 adhering to it, and sprinkles a drop or two from it on 
 to the surface of the water in each cup, so as to cause 
 to sink any minute particles that may happen to remain 
 floating. He then, after an interval of ten minutes, drains 
 off about three-fourths of the supernatant water, which he 
 lets run down the rod into a vessel near him, arid with a 
 tap or two of the rod on the outside of the cup to still 
 further loosen the deposit, this part of the manipulation is 
 concluded. 
 
 The crucibles are next taken to the drying chamber, 
 
 TT 2 
 
092 THE ASSAY OF SILVEE. 
 
 where a steam bath is ready to receive them ; on the per- 
 forated upper plate of this they are ranged, and allowed 
 to remain for about an hour. This gradually, and without 
 spurting, frees the chlorides from moisture, which may be 
 known by their caking, i.e. leaving the sides of the cups 
 round the edges and forming at the bottom of each a 
 loose cake, resembling somewhat a gun-wad. The crucibles 
 are then arranged on a hot-air plate and there exposed to 
 a temperature of between 300 and 350 (F.) for about 
 two hours, till thoroughly dried, when they are ready for 
 weighing.* When the above manipulations have been 
 carefully and satisfactorily gone through, each little cup 
 contains an unbroken, tolerably firm cake of chloride of 
 silver, lying unattached, which admits of being easily 
 grasped with a pair of forceps, and cleanly lifted out of 
 the cup and conveyed to the skiff of the assay balance in 
 which it is weighed. The cups are generally brought 
 from the laboratory to the assayer at the balance in 
 batches of eight or ten. A 6 standard,' synthetically pre- 
 pared of pure silver and copper, and an assay pound of 
 pure silver, are introduced with each day's set of assays, 
 and their chlorides dried with the others, and the analysis 
 of them verified before weighing the rest. Occasionally 
 these ' checks ' are also fused and weighed in a porcelain 
 capsule, but the weight found never differs from that of 
 the chloride merely dried as above. 
 
 Once or twice a month, the silver is recovered from 
 the accumulated chlorides, which are well pounded in a 
 mortar and brought to a powder and then mixed with a 
 proper proportion of chalk and charcoal, and put into 
 a wrought-iron crucible and reduced with heat. The 
 metallic silver so recovered is transferred to the mint. 
 
 Under the circumstances of the solution and of the 
 precipitation as detailed above, should any gold happen 
 to be present in the sample operated on, it is not dissolved, 
 and therefore becomes entangled with the precipitated 
 
 * The chlorides are weighed warm, to obviate the risk of their absorbing 
 moisture ; a precaution especially necessary in the heavy monsoon weather 
 in this country 
 
THE CHLORIDE PROCESS. 693 
 
 silver chloride and dried and weighed with it, and ac- 
 cordingly comes to be regarded and valued as silver. In 
 this the chloride process resembles that by cupellation, 
 which likewise takes no distinguishing cognisance of gold; 
 and both these processes contrast in this respect with the 
 volumetric one, which is a rigid analysis for silver alone ; 
 so that, strictly speaking, an assay conducted by either of 
 the first-named methods ascertains the proportion present 
 of ' the precious metals,' i.e. silver and gold.* 
 
 Should mercury be present it does not interfere with 
 the result, when the solution has been effected in excess 
 of nitric acid with strong heat. Thus the mercury be- 
 comes per oxidised, and hydrochloric acid forms no pre- 
 cipitate in solutions of mercuric salts ; any mercuric 
 chloride resulting from the combination would remain 
 in solution, and be washed away in the course of the 
 process. 
 
 Should lead happen to be present, hydrochloric acid 
 gives no precipitate in a dilute solution, the lead chloride 
 being soluble in a certain proportion of distilled water ; 
 but even were the proportion of lead to silver tolerably 
 large, and the lead chloride happened to be thrown down, 
 the repeated washings would dissolve and get rid of it. 
 
 With regard to the weight of the small portion taken 
 to represent the mass, the system prevails in the Indian 
 mints of taking samples for assay by granulating a small 
 portion of the contents of each melting-pot : when the 
 metal is in a thorough state of fusion and has just been 
 well stirred, a small ladleful of the molten metal is quickly 
 poured from a tolerable height into a vessel of water, and 
 .the granules so formed received on a strainer, lifted out, 
 and perfectly dried. f The weight of this specimen repre- 
 
 * Much of the silver which finds its way to the Indian mints is rich in 
 gold; for instance, sycee contains on an average somewhat about 12 grs. in 
 the troy pound. This in minting operations is considered as silver, and as 
 such it enters into the coinage. There being as yet no refineries established 
 here, through which such silver could pass to the mechanical departments of 
 the mints, the silver coins made during a period when a heavy importation of 
 sycee had been worked up, contain as much as 4 or 6 grains of gold in every 
 32 tolas or 1 pound troy. 
 
 t The introduction into the Calcutta Mint of this system of taking 
 musters is attributable to Dr. Boycott, late Assay Master, and to Dr. Shekelton, 
 
694 THE ASSAY OF SILVER. 
 
 senting each pot was first fixed at 24 grs., technically 
 called the ' assay pound : ' this in the case of pure silver 
 yielded 31-87 grs. of silver chloride, while the same quan- 
 tity of Indian standard silver (which is -j^-ths silver plus 
 T Vth copper= 916-66 in 1000) yielded ^th less, or 29-21 
 grs. : on the weight of chloride being ascertained in 
 each case, a table which was calculated and prepared for 
 the purpose was referred to, and the equivalent fineness 
 assigned to the -Jdwt., plus the odd grs., when any. But 
 when it became desirable to prepare for the decimal form 
 of notation, a number more convenient than 31-97 was 
 looked for to represent purity or 1000, and 25 was fixed 
 on as a desirable starting-point, particularly as the quan- 
 tity of pure silver yielding that amount of chloride, viz. 
 18*825 grs., was quite large enough to represent each pot.* 
 
 The weight, therefore, of the ' assay pound ' in use at 
 present is 18*825 grs. This produces (with chlorine) in 
 the case of pure silver 25 grs. of silver chloride.f 
 
 But to obviate the necessity of constant reference to a 
 calculated table to find the equivalent in pure silver of the 
 amount of silver chloride found in each case, it was inge- 
 niously arranged to stamp each of the assay weights not 
 with its actual weight, but with the figures representing 
 the proportion per mille of pure metal which such a 
 weight of chloride so found corresponds to : thus, sup- 
 
 who, by a number of interesting experiments, satisfied themselves that 
 samples so taken represent the mass of mixed metal to be valued much more 
 fairly than samples of the same mass cut or gouged from it after it has been 
 poured and allowed to cool in the ingot moulds, where a partial separation of 
 the copper from the silver seems to take place ; the result being, according to 
 the above experiments, that in the case of ingots cast in upright moulds all 
 the outside is much below the average fineness of the mass on assay, and the 
 centre much above it. This refers to alloys of silver and copper mixed in or 
 about the proportion of ' standard.' According to M. Levol, however, it would 
 appear that when an alloy of silver and copper in which the proportion of the 
 latter is very high (viz. over 28 per cent.) has been melted, poured, and 
 allowed to cool, an opposite result to the above is found, viz. the outside 
 of the ingots is above the average fineness. An assay, therefore, from a 
 granulated sample must give a much nearer approximation to truth than one 
 from a cut sample. 
 
 * The average weight of the contents of each melting-pot is 12.500 tolas, 
 or about 390 pounds troy, so that the specimen taken to represent this is but 
 about the 119,000th part ; each sample is assayed in duplicate. 
 
 t The basis for these numbers was founded on the proportion in which, 
 according to Turner, silver combines with chlorine, viz. 100 parts with 32'80. 
 
THE CHLORIDE PEOCESS. 695 
 
 posing a melting of five-franc pieces was being assayed, 
 and the chloride resulting from the assay pound operated 
 on weighed 22-5 grs. (showing the actual pure contents in 
 the sample to be 16-94 grs.), instead of referring to a table 
 to see the equivalent per mille-age of silver, that weight 
 which is actually 22*5 grs. has 900 marked on it, and the 
 assayer simply reads the c touch ' from it. 
 
 Accordingly, the assay weights are as follows : 
 
 Actual weight in grs. Figures marked on the weights 
 
 25-000 ....... 1000-00 
 
 22-910 (Std.) 916-66 
 
 22-500 900-00 
 
 20-000 . . ... . . 800-00 
 
 17-500 . . ... . . . 700-00 
 
 15-000 600-00 
 
 12-500 . 500-00 
 
 10-000 400-00 
 
 7-500 . 300-00 
 
 5-000 200-00 
 
 2-500 100-00 
 
 1-250 . ' . .... 50-00 
 
 1-000 40-00 
 
 0-750 30-00 
 
 0-500 20-00 
 
 0-250 10-00 
 
 0-125 5-00 
 
 0-100 4-00 
 
 0-075 . 3-00 
 
 0-050 2-00 
 
 0-025 1-00 
 
 Assay lb., weight=18*825 grs. 
 
 An ordinary day's work consists of eighty assays,* 
 estimating imported bullion to the value of four lacs of 
 rupees, and standard meltings and coins to the value of 
 five lacs. But on emergencies, in time of heavy pressure, 
 by working extra hours, as many as 164 assays have been 
 daily conducted, estimating to the value of eight lacs of 
 rupees, and standard coins and meltings to the value of 
 fourteen lacs. 
 
 Such is an outline of the method of assay worked on a 
 large scale ; of course successful results from it cannot be 
 expected unless each step in the manipulation be con- 
 ducted with great care and accuracy, and only then after 
 much practice and experience. 
 
 * Exclusive of any gold assays which may be going on. 
 
698 THE ASSAY OF SILVER. 
 
 The natives of this country possess great aptitude in 
 acquiring the skill and consequent lightness of touch so 
 essential for delicate manipulation ; this, added to their 
 characteristic patience, makes them admirable subordi- 
 nates in an assay laboratory, under judicious supervision ; 
 moreover, their labour is cheap ; so that, on the whole, 
 the process seems to be especially suitable for an Indian 
 mint. 
 
 When bar silver is imported from the Continent, the 
 assays of it, made" here, almost invariably correspond most 
 closely with those previously made of it in Paris by the 
 volumetric method. But were further proof needed of 
 the practical accuracy of this system, it is to be found in 
 the very close proximity to the legal standard at which 
 the large Indian coinage has been maintained for many 
 years, as annually reported by the assayers to the Eoyal 
 Mint of Great Britain, who test the fineness of the Indian 
 pyx coins by the French process. 
 
 Without this me tBod (improved and made more per- 
 fect, as it has been, in the hands of successive assay 
 officers), it would have been very difficult for the assay 
 establishments of the Indian mints to have dealt with, in 
 the same time and with the same accuracy, the immense 
 importation of silver to India during "the last fifteen years. 
 In the single year 1865-66 there was poured into the 
 Indian mints, and manufactured into coin, silver alone 
 reaching in value to the prodigious amount of over four- 
 teen millions sterling. 
 
 The system which enabled the assay officers to value 
 such a rapid and heavy influx with accuracy, and with 
 satisfaction to the importer on the one hand, and to the 
 mint (the buyer) on the other, and to faithfully maintain 
 the immense resulting coinage close to legal standard, has 
 been put to a severe test. If success be the criterion of 
 merit, the twenty years' large experience of this method 
 gained in the Indian mints goes to show that it is worthy 
 of a yet wider field of utility. 
 
THE CHLORIDE PROCESS. 697 
 
 Apparatus and Appliances required. 
 
 (1.) The bottles used in this process are of thin (but 
 strong) white glass, and contain about 12 fluid oz. ; about 
 6 inches in height and 2^ inches in diameter at the bottom, 
 which should present a perfectly even, level floor ; they 
 are without any abrupt shoulder, but become gradually 
 pyramidal from about half-way up to the neck ; this shape 
 favours the easy dropping out of the chloride. The neck 
 is about one inch in length, polished on its inner surface ; 
 the stoppers are of ground glass, polished, with globular 
 heads, and are made to fit with the utmost accuracy and 
 smoothness. The bottles and stoppers are numbered, to 
 correspond with the number on the muster board and also 
 on the cups. 
 
 (2.) The ' cups' are Wedgwood crucibles, smooth and 
 thin, about 1^ inches in height, 1^ inches in diameter 
 above, and a little less than 1 inch in outside diameter at 
 the bottom. The floor should be perfectly level, and 
 neither it nor the sides should present any roughness likely 
 to retain the chloride. The cups are all numbered. 
 
 (3.) The porcelain saucers are shallow, f of an inch in 
 depth ; the upper diameter is about 4 inches, the lower 2-J 
 inches. 
 
 (4.) The turn-table is a circular board of about 3 feet 
 in diameter, fenced by a brass railing (or by a simple ledge) ; 
 its centre is occupied by a raised platform about 2 feet in 
 diameter, between which and the rail the bottles (20 on 
 each) stand, the round outer edge of the platform having 
 semi-lunar niches cut in it, into which the bottles fit ; 
 opposite to each niche on the platform is a little concavity 
 in which the stoppers rest when not in the bottles. Each 
 turn-table is made to revolve on its centre in either direc- 
 tion, and is raised about 6 inches above the long general 
 table on which all are supported ; close to each a funnel 
 is fitted into the lower (supporting) table for conducting 
 away the fluid syphoned from each set of bottles. 
 
 (5.) The trough is a basin of cast iron (painted) ; it 
 
698 THE ASSAY OF SILVER. 
 
 may be oblong or round, raised to about the height of 
 3 feet from the ground ; when round and large enough 
 for twenty bottles, space and distilled water may be 
 economised by having a platform in 'the centre. This is 
 convenient for resting the bottles oh after the chlorides 
 have been got out. A trough of this kind may 
 be about 2^ feet in diameter, having a space 7 inches 
 broad and 4 deep all round between the circumference of 
 the basin itself and the outer edge of the island platform. 
 Into this space is poured distilled water to the depth of 
 3 inches. From the rim of the trough hang as many 
 brass supports as there are bottles to be inverted ; there 
 are two circular clasps connected at the back to a bar 
 common to both ; one, the larger, is 1-| inch above the 
 smaller and lower one, which is under water ; they are 
 open in front (or towards the centre of the basin) to about 
 J of an inch in width. The openings of both are in the 
 same line, owing to the lower (smaller) segment being 
 projected towards the centre by an abrupt curve in the 
 connecting bar, by which they hang from the brim. This 
 arrangement receives and fixes the inverted bottles in the 
 required position. The distilled water is removed from 
 the trough by the withdrawal of a plug. These troughs 
 are sometimes made to revolve on the centre. 
 
 (6.) The drop bottle used for washing down the glass 
 rod when breaking up the chlorides, and for sprinkling the 
 surface of the cups, is small-sized, round, so as to be easily 
 grasped ; it holds about 6 oz. The stopper is hollow, 
 with two small tubes leading from its head, one opposite to 
 the other. Glass is so liable to break or chip, that a hollow 
 silver stopper is now generally substituted. 
 
 (7.) The steam-bath is simply a square vessel made 
 of sheet copper, between 3 and 4 inches deep, the top or 
 upper plate of which has a number of circular openings 
 about two-thirds of the diameter of a Wedgwood crucible. 
 There is also a steam escape pipe leading from the centre 
 below to about a foot in height. They are of various sizes, 
 to contain from 10 to 150 pots : they are raised or moved 
 by two lateral handles. 
 
EFFECT OF BISMUTH ON SILVER. G99 
 
 (8.) Hot-air plate of thin sheet iron bored with holes 
 for the reception of the crucibles, raised by iron feet about 
 1-| inch above the furnace plate. It is furnished with a 
 square tin cover, which fits over it. This is provided with 
 lateral apertures for the escape of heated air, and with a 
 tube from its roof for the reception of a thermometer. 
 
 The drying furnace on which the above rest is sur- 
 mounted by a hood, the glazed door of which slides up 
 and down by weights and pulleys ; the plate is heated by 
 means of gas jets ; it has a good draught to carry off the 
 nitrous fumes, as on it the musters are dissolved in the first 
 instance on a sand bath. 
 
 (9.) The forceps for removing the cake of chloride 
 from each cup to the skiff of the balance should not be 
 too sharp in its grasp ; it is much improved by having 
 the blades tipped for about an inch from the points with 
 platinum about ^ inch in width. 
 
 (10.) It is a convenience to have the assay weights 
 arranged in a set of ivory compartments in the weight 
 box ; on the floor of each compartment are engraved the 
 figures corresponding to those engraved on the weight 
 which occupies it ; by this means the assayer has merely 
 to glance at his weight box to see what weights are in the 
 pan of the balance, and to read off the ' touch ' when each 
 chloride is counterpoised. 
 
 Effect of Bismuth on the Ductility of Silver. 
 
 The effects produced by small quantities of bismuth on 
 the ductility of silver have been carefully investigated by 
 Surgeon-Major J. Scully, Assay Master, Calcutta. 
 
 The following are extracts from an elaborate paper 
 which he published in the Chemical News for 1887 : 
 
 It is well known that alloys of silver and bismuth, in 
 certain proportions, are brittle. In Dr. Percy's valuable 
 work on Metallurgy (Silver and Gold Part I.) it is stated 
 that alloys of silver w^ith bismuth, in the proportion of 
 50 per cent, and 33 per cent, of the latter metal, are 
 brittle ; while an ore of silver and bismuth, called 
 
700 THE ASSAY OF SILVER. 
 
 Chilenite, in which bismuth occurs only to the extent of 
 14*4 to 15*3 per cent., is said to be malleable. The least 
 amount of bismuth, however, which will injuriously affect 
 the ductility of silver, for example, in such an operation 
 as the lamination of silver bars for coinage, does not, so 
 far as I am aware, appear to have been experimentally 
 investigated. It may here be mentioned at the outset 
 that an alloy of silver and bismuth may, by careful 
 hammering, be extended considerably, so as to pass muster 
 as malleable ; although, if subjected to lamination by 
 means of steel rolls, the same alloy will crack at the 
 edges and thus show a deficient ductility, as compared 
 with pure silver or some silver-copper alloys. It is to the 
 deficiency in ductility, as tested by rolling, of silver con- 
 taining only very small proportions of bismuth that I here 
 wish to call attention. 
 
 My attention was first prominently directed, about two 
 years ago, to the injurious effects caused by small quan- 
 tities of bismuth in silver by the circumstance that some 
 silver bullion, in the shape of English refined bars of as 
 high a fineness as 990 per mille, proved so brittle as to be 
 unfit for mintage. Attention was first attracted to this 
 matter by the peculiar behaviour under assay of the 
 granulated samples taken from this silver after melting. 
 The appearances noticed under assay will be referred to 
 presently, but they led to the bullion being at once tested 
 for brittleness. A bar, about 21 inches long, 2-25 broad, 
 and 1 inch thick, was hammered out at one end without 
 cracking, but on being passed through the rolls it cracked 
 badly at the edges and was pronounced to be ' brittle.' 
 in the Mint sense of the term. The bullion was then re- 
 melted in five plumbago pots, and a partial refinement of 
 it attempted in the ordinary way with nitre, about 8 to 
 10 Ibs. of this salt being used for each pot. The resulting 
 silver bars were not appreciably improved by this treat- 
 ment ; hammering again proved an inconclusive test, but 
 a bar of the size I have mentioned broke in two by merely 
 dropping on the floor of the melting-room. 
 
 In the meantime the assay had shown that the brittle 
 
EFFECT OF BISMUTH OS SILVER. 701 
 
 bullion contained bismuth, and that this was the only 
 substance present likely to be the cause of brittleness. 
 The Indian process of assaying silver has already been 
 described (p. 688) by Dr. Busteed. The main features of 
 the process may here be briefly recapitulated for the 
 purpose I have in view. 
 
 A fixed weight of the silver bullion to be assayed is dis- 
 solved in an assay bottle, by means of nitric acid aided by 
 heat ; the solution is diluted with water and an excess of 
 hydrochloric acid is added, to precipitate all the silver pre- 
 sent as chloride. The silver chloride having been caused 
 to aggregate and settle by vigorous shaking, the bottle is 
 filled up with water, and the supernatant fluid is subse- 
 quently syphoned off, to remove all the now dissolved 
 matter which may have been contained in the bullion. 
 
 Under these conditions of solution, precipitation, and 
 dilution with water, chemists will readily understand that 
 even a small trace of bismuth, if it be in the silver, will 
 reveal its presence by the partial formation of insoluble 
 oxy chloride of bismuth. Now, in the assay of the brittle 
 bullion under consideration, solution in nitric acid had 
 been readily and completely effected by the aid of heat ; 
 antimony and tin were consequently absent. After the 
 addition of water and hydrochloric acid, however, the 
 solution in the assay bottles could not be cleared by 
 shaking ; the bulk of the silver chloride collected at the 
 bottom of the bottles, but the supernatant fluid remained 
 turbid. 
 
 Tin and antimony being excluded, only two metals 
 could produce this result in the wet assay of silver 
 namely, mercury and bismuth. To determine which of 
 these is the interfering metal it is only necessary to note 
 the effect of solar light on the silver chloride formed ; 
 when mercury is present the silver chloride maintains its 
 pure white colour unaltered, while in the presence of 
 bismuth the chloride immediately acquires the well-known 
 purple colour under the influence of daylight. Our 
 assays, then, being turbid after precipitation and yet the 
 silver chloride blackening readily under the influence of 
 
702 THE ASSAY OF SILVER. 
 
 daylight, it was evident that bismuth was present. The 
 turbidity produced was due to the partial formation of 
 bismuth oxychloride ; and this compound diffusing itself 
 in its characteristic manner through the solution had 
 broken up part of the silver salt into very fine powder, so 
 that some hours had to elapse before the supernatant fluid 
 cleared by the gradual subsidence of both bismuth oxy- 
 chloride and the finely divided silver chloride. The assay 
 was of course thus rendered unreliable, since the silver 
 chloride to be weighed, and on which the calculation of 
 the fineness rested, was contaminated with bismuth oxy- 
 chloride. A cupellation assay of this bullion was at once 
 had recourse to for ascertaining its fineness. 
 
 So far, then, this tender of silver bullion seemed to 
 establish the following points: 
 
 1. Silver bullion of as high fineness as 990 per mille 
 is rendered unfit for coinage purposes by an amount of 
 bismuth which, in this particular case, could not possibly 
 have exceeded 1 per cent., and was probably less than that 
 proportion. 
 
 2. Hammering a bar of silver bullion is not a good 
 test for detecting brittleness, as far as Mint purposes are 
 concerned. 
 
 3. The toughening of silver bullion 990 fine, and con- 
 taining only a small amount of bismuth, by the aid of 
 nitre in plumbago melting-pots, is not readily affected. 
 
 4. The presence of a trace of bismuth in silver of 
 high fineness is immediately detected in the ordinary course 
 of assay by the Indian method, but this bismuth interferes 
 with the perfect accuracy of the results obtained by that 
 process. 
 
 A comprehensive research seemed, therefore, called for 
 to elucidate the whole subject, and the necessity for this 
 investigation has since been emphasised by the fact that 
 silver bullion contaminated with bismuth has frequently 
 found its way to the Mint since its first discovery here. 
 The points to be investigated seemed naturally to group 
 themselves under the following heads : 
 
 1. Is our ordinary wet assay of silver susceptible of 
 
EFFECT OF BISMUTH ON STLVEE. 703 
 
 such easy modification as will enable us to obtain perfectly 
 accurate results by it, in presence of bismuth, without 
 having recourse to the confessedly less accurate assay by 
 cupellation ? And, how may small quantities of bismuth 
 in silver be readily estimated with the despatch indispens- 
 able for mint operations ? 
 
 H. What is the smallest amount of bismuth in silver 
 that will render it unfit for coinage, when present in bars 
 of the Indian standard fineness of 916*6 ? And, what is 
 the amount of bismuth that may be tolerated in such bars 
 without materially injuring the ductility ? 
 
 III. How is silver bullion containing bismuth, which 
 may be tendered to the Mint, to be dealt with, supposing 
 that establishment accepts any metal that is brittle ; and 
 how is the presence of bismuth in refined bars to be 
 accounted for ? 
 
 1. As the purity of the bismuth to be used in the 
 experiments now to be detailed was a matter of first im- 
 portance, I may briefly mention the steps taken to insure 
 the purity of the metal. Eefined bismuth was dissolved 
 in nitric acid, precipitated as basic nitrate by diluting 
 largely with distilled water, the nitrate digested in solu- 
 tion of caustic potash, and then well washed, dried, and 
 reduced by heating with charcoal in a clay crucible. A 
 series of synthetical assays, made by dissolving together 
 pure silver and pure bismuth, the latter in the proportion 
 of from 1 to 5 thousandths, showed that our ordinary 
 process of assay, under such conditions, gave unreliable 
 results, there being a surcharge, or higher report than 
 should have been obtained, which varied from 0'7 to 2- 7 
 mils., when the proportion of bismuth was from 3 to 5 
 thousandths. A modification in our process of assay was 
 evidently required if it were to be used for estimating the 
 fineness of silver bullion containing bismuth ; and the neces- 
 sary steps to this end were, after repeated experiment, 
 found to consist in adding the smallest possible amount of 
 hydrochloric acid for the precipitation of the silver, and 
 increasing the amount of nitric acid in which it was first 
 dissolved. We use ordinarily for the precipitation of an 
 
704 THE ASSAY OF SILVER. 
 
 assay pound of silver 5-4 c.c. of hydrochloric acid, of 
 sp. gr. 1075 ; but 2*5 c.c. of acid of this strength suffices 
 for the complete precipitation of an assay pound of even 
 fine silver ; so that we have here at once a means of 
 diminishing the tendency of any bismuth in the silver to 
 form insoluble oxy chloride. If, in addition to diminishing 
 the amount of hydrochloric acid, we added a considerable 
 excess of nitric acid to the solution (which acid would 
 not in any way interfere with the silver chloride formed), 
 all risk of the partial formation of insoluble bismuth salts 
 seemed removed. This in fact has proved to be the case, 
 and the successful modified process for the assay of silver 
 containing bismuth is as follows : 
 
 The assay pound of silver bullion containing 'bismuth 
 is dissolved in 5-5 c.c. of nitric acid, sp. gr. 1200, with 
 the aid of heat, about 5 ounces of water are added, 
 and then 10 c.c. of nitric acid, sp. gr. 1320. The silver 
 is now precipitated by the addition of 2-5 c.c. of hydro- 
 chloric acid, and after vigorous shaking the supernatant 
 fluid will be found perfectly clear, and it will remain so 
 when the bottle is filled up with water, all the bismuth 
 present being in solution. Whenever samples of silver 
 now show the presence of bismuth during the assay, a 
 fresh set is taken up and worked by the modified process, 
 the delay thus caused not amounting to more than a 
 few minutes. It may be mentioned here that all our 
 assays are reported to one-tenth of a millieme (0*1 per 
 mille). 
 
 Having thus ascertained the presence of bismuth in 
 silver bullion, and put in practice a modification of the 
 assay process which renders us indifferent to its presence, 
 it is still of importance to ascertain the exact proportion 
 of bismuth which is present in the bullion, and, to be of 
 practical use for mint work, this estimation must be 
 effected rapidly and as simply as possible. The ordinary 
 directions given for the separation of bismuth in the pre- 
 sence of silver, by first removing the latter as chloride 
 and then precipitating the bismuth as carbonate, do not, 
 I find, give accurate results when silver is present in such 
 
EFFECT OF BISMUTH ON SILVEK. 705 
 
 overwhelming proportions as obtain in the cases under 
 consideration. 
 
 I have therefore adopted the following plan, which a 
 number of synthetically prepared solutions have proved 
 to give quick and good results, though sometimes the 
 amount of bismuth present is very slightly underestimated. 
 The ordinary silver assay having given a rough visual 
 estimate of the amount of bismuth likely to be present, 
 enough of the bullion is taken to yield a fairly weighable 
 amount of bismuth oxide in the final result. The bullion 
 is dissolved in a small amount of nitric acid, the solution 
 carefully diluted, and an excess of ammonium carbonate 
 at once added, the precipitation being aided by heating. 
 The carbonates of silver and copper at first formed are 
 re-dissolved, and the carbonate of bismuth after a time 
 settles completely at the bottom of the beaker. The con- 
 tents of the beaker are then passed through a filter, of 
 which the weight of ash yielded by incineration is known, 
 and the carbonate of bismuth on the filter washed quite 
 free of all traces of silver. The filter is then dried, its 
 contents transferred to a porcelain crucible for ignition, 
 the filter-paper being ignited separately, treated with a 
 drop or two of nitric acid to re-oxidise any bismuth oxide 
 reduced by contact with the carbon of the filter, and the 
 ash added to the crucible. From the weight of bismuth 
 oxide thus found, after deducting the weight of the filter 
 ash, the amount of metallic bismuth present in the sample 
 of bullion taken for analysis can be at once found. 
 
 There are only two metals likely to interfere with the 
 accuracy of the process here described namely, cadmium 
 and lead ; the carbonates of both these metals being as 
 insoluble in excess of the precipitant employed as bismuth 
 carbonate. Cadmium is very unlikely to be found in 
 silver bullion and its consideration may be neglected, but 
 if the presence of lead is suspected the carbonate filtered 
 from the silver solution is dissolved in nitric acid, evapo- 
 rated down with the addition of sulphuric acid, and the 
 lead sulphate formed (if any) collected and weighed in 
 the usual way. The bismuth is again precipitated as 
 
 z z 
 
706 THE ASSAY OF SILVER. 
 
 carbonate and treated as before directed. Many experi- 
 ments have been made with synthetically prepared mix- 
 tures of silver, copper, lead, and bismuth, the latter two 
 metals being in very small proportion to the silver, so as 
 to imitate the composition of some refined bars. Ullgreen's 
 plan for the separation of the carbonates of lead and 
 bismuth, by dissolving them in acetic acid and then pre- 
 cipitating the bismuth by means of a lead rod, does not- 
 work satisfactorily and requires too long a time for the 
 precipitation. 
 
 II. As it seemed likely that a large number of ex- 
 periments would be required to estimate accurately the 
 smallest amount of bismuth that would injure the ductility 
 of our coinage alloy, and the still smaller proportion that 
 would not sensibly affect this ductility, it was determined 
 to begin the inquiry by a number of laboratory experi- 
 ments on small bars of silver, before trying the effects of 
 bismuth on ordinary coinage bars and with the procedure 
 for lamination carried out in the Mint. These laboratory 
 experiments were made in the following way : Pure 
 silver prepared for assay check purposes, or an alloy of 
 silver and copper of which the exact composition had 
 been estimated by assay, was melted in a clean plum- 
 bago crucible under charcoal. When the metal was in 
 fusion the necessary amount of bismuth was rolled in a 
 piece of paper, carried down at once to the bottom of the 
 silver bath, and then thoroughly mixed with the silver by 
 stirring. The calculated composition was confirmed by 
 assay of the silver. When this mixture had been accom- 
 plished, the contents of the crucible were poured into an 
 open iron ingot mould, and after cooling either quickly 
 by plunging the casting into water or slowly in contact 
 with the mould the bar so cast was tested for brittleness 
 by hammering and rolling. The bars cast were of two 
 sizes, one set being 3'75 inches long, 1-125 broad, 0-375 
 thick, and weighing about 6*2 troy ounces ; and another 
 set 2-69 inches long, 1-125 broad, 0-25 thick, and weigh- 
 ing about 4-1 troy ounces. When reduced to the fullest 
 extent by rolling, these bars were converted into straps 
 
EFFECT OF BISMUTH ON SILVER. 707 
 
 about 0/015 inch in thickness. In laminating them they 
 were twice annealed, first after having undergone four 
 pinches in the rollers, and again after the tenth pinch 
 from the beginning. Similarly shaped bars of silver, 
 without bismuth, were occasionally laminated in the same 
 way to obtain a sure means of comparison. Before any 
 result was accepted as to brittleness or its absence, the 
 bar under experiment was always re-melted and tried at 
 least a second time. The number of experiments in this 
 series amounted to fifty -three, and the following is a 
 summary of the results obtained : 
 
 Fine silver when alloyed with only 1 per mille (one- 
 thousandth part of its weight) of bismuth, and the casting 
 rapidly cooled by plunging it into water as soon as it has 
 set, has its ductility, as tested by lamination, sensibly but 
 slightly impaired, the straps resulting from rolling having 
 slightly jagged edges. When the proportion of bismuth 
 is increased to 2, 3, 4, and 5 per mille, the plan of cooling 
 remaining the same, the raggedness of the edges of the 
 straps was somewhat increased, but not very markedly. 
 If, however, the casting was allowed to cool down com- 
 pletely, but very slowly, in contact either with the mould 
 or a stone floor, the results were very different. Under 
 this condition of cooling, a bar composed of fine silver 
 with 4 per mille of bismuth was completely brittle ; it was 
 readily broken, and its fracture was strongly crystalline. 
 On laminating it small cracks appeared all over the sur- 
 face on the second pinch, the bar emitting a crackling 
 sound under the rolls, much like the ' cry ' of tin, and on 
 the fourth pinch the bar cracked deeply at the edges. 
 This remarkable effect on the molecular structure of this 
 alloy of silver and bismuth, as due solely to the mode of 
 cooling the casting, was repeatedly verified on the same 
 metal by re -melting and cooling rapidly and slowly alter- 
 nately. The case seems analogous to that of bronze, 
 where slow cooling of the alloy after casting is said to 
 make it hard and brittle. 
 
 Fine silver with 6 per mille of bismuth, rapidly cooled, 
 was distinctly cold-short and crystalline on fracture ; the 
 
 z 2 
 
708 THE ASSAY OF SILVEK. 
 
 bar cracked on the surface at the fourth pinch. With 
 7 per mille of bismuth these evidences of diminished 
 ductility were slightly more pronounced. With 8 per 
 mille of bismuth the silver was still more brittle, the bar 
 broke readily when hammered, and cracked all over the 
 surface on the fourth pinch from the rolls. With 9, 10 r 
 and 11 per mille of bismuth, the bar of silver could be 
 readily broken in two by merely striking it against the 
 edge of an anvil, the fracture was coarsely crystalline, and 
 the bar, in one case, proved to be very red-short, a mere 
 tap from the tongs sufficing to break it in two when 
 heated for the purpose of annealing. Although these bars 
 were so very brittle, it was still possible to roll them into 
 thin straps after careful annealing ; but the edges of the 
 straps so produced were deeply jagged and indented by 
 cracks. These bars also all emitted the peculiar crack- 
 ling noise under the rolls which has before been men- 
 tioned. 
 
 An alloy containing 990 parts of silver and 10 of 
 copper then had added to it successively 1, 2, 3, 4, and 5 
 per mille of bismuth, the castings being rapidly cooled. 
 The remarks already made with reference to fine silver 
 alloyed with the same proportions of bismuth would 
 apply here almost exactly that is to say, the bars were 
 rolled out to a thickness of 0-015 with somewhat ragged 
 edges, so that although ductility, as thus tested, was 
 impaired, it was only slightly so. With 6 per mille of 
 bismuth (fineness of metal on assay 983-9) the edges 
 cracked a little, and, after annealing and rolling out, the 
 strap had decidedly jagged edges and was split for some 
 distance at one end. The bars containing 4, 5, and 6 per 
 mille of bismuth were now re-melted and allowed to cool 
 slowly and completely in the mould. They were all found 
 to be highly brittle, broke easily under the hammer 
 the fracture being granular and not crystalline and on 
 being rolled they cracked badly, all over the surface and 
 at the edges, on the first or second pinch ; in one case the 
 bar broke in two on the second pinch. That these very 
 different results were again solely due to the manner of 
 
EFFECT OF BISMUTH ON SILVER. 709 
 
 cooling was proved by re-melting and rapidly cooling the 
 castings, when the same metal proved comparatively 
 ductile, as first stated. 
 
 Silver of the Indian standard of 916*6 per mille (the 
 rest being copper) to which 2 per mille of bismuth was 
 added, gave on lamination straps with slightly jagged 
 edges and proved to be red-short. With 4 per mille of 
 bismuth the bars showed a few surface cracks on being 
 rolled, and the resulting straps had decidedly jagged 
 edges. Slow cooling of these castings did not affect their 
 ductility, thus showing a marked contrast to what had 
 been observed in the case of fine silver and the alloy con- 
 taining only 10 per mille of copper. When the amount 
 of bismuth was increased to 5 per mille, the copper 
 present remaining at 834 per mille, the bars were de- 
 cidedly brittle and cracked readily on hammering the 
 fracture being again granular, and not crystalline as in 
 the case of fine silver. On lamination both surface and 
 edge cracks developed after four pinches from the rolls, 
 and in annealing one of these bars the whole surface 
 blistered considerably, no doubt owing to the temperature 
 having been carried a little too high. Standard silver 
 with 10 per mille of bismuth, reducing the fineness as 
 ascertained by assay to 906*6, was very brittle, the bars 
 breaking easily under the hammer, and on the fourth 
 pinch from the rolls splitting and cracking all over the 
 surface. In the course of these latter experiments it was 
 ascertained that, with from 83-5 to 70 per mille of copper 
 present, slow or rapid cooling of silver alloys containing 
 bismuth made no appreciable difference in their ductility. 
 
 The foregoing experiments having furnished some in- 
 formation as to the amount of bismuth that might be 
 expected to injure our coinage alloy, it was now decided 
 to test that point practically, by operating on coinage bars 
 subjected to the regular procedure for the manufacture of 
 rupees in the Calcutta Mint. The experiments made in 
 this connection were fourteen in number. The bars used 
 here for coinage weigh about 253 troy ounces, and are 
 about 20 inches long, 2-25 broad, and 1 inch thick ; they 
 
710 THE ASSAY OF SILVER. 
 
 are cast in vertical iron moulds. In lamination they are 
 first reduced by eleven pinches to a thickness of 0'23 in. ; 
 they are then annealed, and finally reduced by twelve 
 additional pinches to a thickness of 0*06 inch. A number 
 of bars, poured from a pot of which the contents had 
 proved on assay of a granulated sample to be 916-6 fine, 
 were selected for the experiments, and as a preliminary 
 step one of the bars was laminated to test its ductility, 
 It rolled out with smooth ' wire ' edges, and indeed its 
 ductility was beyond suspicion, as it resulted from a 
 melting of good coins. Another bar of the same batch 
 was now melted and 1 per mille of bismuth added to it, 
 the result of the addition being checked, in this and all 
 following cases, by the assay of a granulated sample of 
 the metal, taken after thorough stirring. At the eighth 
 pinch both edges of the lower half of this bar began to 
 crack, and at the eleventh pinch these cracks extended 
 towards the middle line of the strap for about a quarter 
 of an inch, and occurred at about every half inch of the 
 edge. After annealing, and in the subsequent lamination 
 to a thickness of 0-065 inch, these cracks increased con- 
 siderably in number, but did not become sensibly deeper. 
 The strap as finished was pronounced unfit for coinage 
 purposes ; for although two blanks could have been cut 
 from its width, the edges were too jagged to admit of the 
 blanks being obtained exactly along the line from which 
 it was desired to cut them this position being attained 
 by means of a fixed lateral guide against which the edge 
 of the strap had to be maintained in cutting. With 2 per 
 mille of bismuth the results obtained on rolling were not 
 much worse than with 1 per mille. But the side cracks 
 opened out more, and here again it was noticed that the 
 lower portion of the bar (upper and lower here having 
 reference to the casting in upright moulds) was somewhat 
 less ductile than the upper part. 
 
 With 3 per mille of bismuth (fineness on assay 913'8) 
 the bar began to crack on both edges at the ninth pinch ; 
 at the eleventh pinch there were many cracks quite a 
 quarter of an inch deep, and after annealing the bar these 
 
TITRATION OF SILVER IN PRESENCE OF COPPER, ETC. 711 
 
 cracks increased at every pinch, so that at the twenty-first 
 pinch the strap was cracked all along both edges very 
 badly. It would only have been possible to obtain one 
 blank from the width of this strap. 
 
 As it was perfectly clear that no further experiments 
 were required with larger proportions of bismuth, the 
 subsequent trials were made on coinage bars containing 
 0-5 per mille, 0-25 per mille, and, by dilution of the latter 
 bars with standard silver, to even half and a quarter of 
 the lesser proportion just stated. Here the results were 
 rather discordant ; they appear to have been somewhat 
 influenced by the state of different rolls, and by quick or 
 slow annealing. The general outcome of the tests, how- 
 ever, was that although some of the straps containing the 
 proportions given of bismuth were jagged at the edges, 
 and so would have yielded a diminished percentage in 
 outturn of good blanks, others were not materially worse 
 than the average of straps without any bismuth at all. 
 As a result of this part of the inquiry, it may, I think, be 
 fairly concluded that if our coinage bars contain less than 
 0-5 per mille of bismuth their ductility will not be mate- 
 rially affected. It must be borne in mind that these results 
 only apply to bars of the size and shape of those experi- 
 mented on, and with the particular treatment in lamina- 
 tion above detailed. With thinner bars and a different 
 method of rolling, different results may be expected. The 
 system of cutting out blanks has also to be considered, 
 for in some mints straps with saw edges are not so pre- 
 judicial as in others. 
 
 Titration of Silver in Presence of other Metals by means of 
 Ammonium Sulpho cyanide. 
 
 Mr. J. Volhard gives the following process for the 
 volumetric assay of silver. 
 
 In a nitric solution of silver to which a soluble ferric, 
 salt has been added, a permanent redness does not appear 
 on the gradual addition of a dilute solution of ammonium 
 or potassium sulphocyanide, until all the silver has been, 
 
712 THE ASSAY OF SILVER. 
 
 thrown down as a sulphocyanide. If we know how much 
 of the sulphocyanide solution is required to precipitate a 
 known weight of silver we can estimate volumetrically 
 the quantity of silver present in any acid argentic solution, 
 the ferric salt serving as an indicator. For the prepara- 
 tion of the standard solution the author uses ammonium 
 sulphocyanide, though Lindermann prefers the potassium 
 salt ; both in dilute solution are permanent. But the am- 
 monium salt is less frequently contaminated with chlorides, 
 which interfere greatly if present in more than mere traces. 
 The solution may be conveniently made of such a strength 
 that 1 c.c. indicates 1 centigramme of silver. The weighed 
 quantity of the salt is dissolved in water, and diluted in a 
 graduated flask so far that 7*5 grins, (or 8 grins, if the 
 salt appears very damp) may be contained in each litre. 
 For the precipitation of 1 grm. of silver 0-704 grm. of 
 pure ammonium sulphocyanide is requisite ; 0'5 grm. of 
 chemically pure silver is then weighed out, dissolved in 8 
 or 10 c.c. of pure nitric acid of sp. gr. 1*20, and after the 
 complete solution of the metal the nitrous acid is expelled 
 by boiling or prolonged heating on the sand-bath, and the 
 solution is allowed to cool. It is then diluted with 200 
 c.c. of water, and 5 c.c. of a cold saturated solution of 
 ammonia-iron alum are added. If the colour of the ferric 
 salt is perceptible, a little pure colourless nitric acid is 
 added till it disappears. The sulphocyanide solution is 
 then added from a burette. At first a white precipitate is 
 produced, which remains suspended in the liquid like silver 
 chloride, rendering it milky. On the further addition of 
 sulphocyanide each drop produces a blood-red cloud, which 
 quickly disappears on agitation. As the point of satura- 
 tion is reached, the silver sulphocyanide collects in flocks, 
 and the liquid grows clearer, without becoming perfectly 
 limpid, as long as a trace of silver remains in solution. As 
 soon as all the silver is precipitated the flocculent precipi- 
 tate quickly deposits, and the supernatant liquid becomes 
 quite clear. The sulphocyanide solution is added by drops 
 till this point is attained, and till a very faint light-brown 
 colour appears in the liquid, which does not vanish on 
 
TITRATION OF SILVER IN PBBSBNCB OF COPPER, ETC 713 
 
 repeated agitation. The colour is most easily perceived 
 if the liquid is held, not up to the light, but against a white 
 wall turned away from the window. 
 
 For a repetition of the experiment it is convenient to 
 use a silver solution of a known strength. For this pur- 
 pose 10 grms. of pure silver are dissolved in nitric acid 
 in a litre flask, the nitrous acid is expelled, the liquid is 
 allowed to cool, and diluted up to the volume. For use 
 50 c.c. are taken with a pipette. If the above-mentioned 
 proportions are preserved for 0'5 grin, silver, or 50 c.c. of 
 the silver solution, somewhat less than 50 c.c. of the sul- 
 phocyanide solution will be required, whence the needful 
 dilution can be calculated. If 48*5 c.c. sulphocyanide 
 have been used, then to every 48-5 c.c. of the solution 15 
 c.c. of water must be added. 
 
 The solution is perfectly permanent, even on being 
 kept for two years. The silver solution must have a 
 decidedly acid reaction from free nitric acid, which it is 
 unnecessary to neutralise, although it is important that 
 the proportion of acid in different experiments should be 
 approximately equal. It is, however, essential for the con- 
 stancy of the results that the ferric salt should be in large 
 excess, and should be used approximately in one and the 
 same proportion to the total volume of the fluid. It must 
 also be remembered that the colour of ferric sulphocyanide 
 is destroyed by nitrous acid at common temperatures, and 
 by nitric acid on the application of heat. Hence follows 
 the necessity of completely expelling all nitrous acid, and 
 of allowing the liquid to become cold before the operation 
 is begun. The nitric acid used in this process should be 
 kept in the dark. 
 
 Copper. The proportion of copper in an alloy may 
 reach 70 per cent, without in the least affecting the accu- 
 racy of the process. Beyond this limit the recognition of 
 the final reaction is somewhat doubtful, since after the 
 precipitation of the silver the liquid is rendered opaque 
 and discoloured by the formation of black copper sulpho- 
 cyanide. The only remedy in such cases is to add a 
 known weight of pure silver to the alloy, so as to reduce 
 
711 THE ASSAY OF SILVER. 
 
 the proportion of copper below 70 per cent., as is done in 
 Gay-Lussac's process. 
 
 Mercury. In presence of this metal silver cannot be 
 titrated with sulphocyanide solution. This defect is of 
 little consequence, as mercury is readily expelled. 
 
 Palladium renders the titration of silver with sulpho- 
 cyanide inaccurate, this metal appearing in the result as 
 silver. This is an important circumstance from a technical 
 point of view, since palladium, though frequently present 
 in silver, occurs in very small quantities. 
 
 As regards other metals soluble in nitric acid and 
 found along with silver in alloys and ores, lead, cadmium, 
 thallium, bismuth, zinc, iron, and manganese are without 
 influence. The recognition of the final reaction in solu- 
 tions strongly coloured by salts of cobalt or nickel, 
 requires some practice. At first a few drops of sulpho- 
 cyanide will always be added in excess. It is then recom- 
 mended to titrate back with a solution of silver, when the 
 pure colour of the cobalt or nickel will appear so suddenly 
 and distinctly, that it will conversely be easy to seize the 
 exact point, when the colour of the solution is modified 
 to a yellowish brown by the addition of the colour of fer- 
 ric sulphocyanide. When the change of shade has been 
 observed four or five times by titrating backwards and 
 forwards with exactly corresponding solutions of silver 
 and of sulphocyanide, such a certainty in the recognition 
 of the final reaction will be attained that there can 
 be no doubt as to the addition of a half drop more or 
 less. 
 
 The presence of tin, antimony, and arsenic does not 
 interfere with the accuracy of the process. 
 
 Mr. C. A. M. Balling gives the following instructions 
 for the direct estimation of silver in galena by Vol- 
 hard's process : From 2 to 5 grins, of the galena, accord- 
 ing to its supposed richness in silver, are very finely ground 
 and intimately mixed in a porcelain mortar with from 
 three to four times its weight of a flux composed of equal 
 parts of soda and saltpetre, placed in a porcelain crucible, 
 covered, and heated over a burner to thorough fusion, 
 
BLOWPIPE ASSAY OF SILVER. 715 
 
 when the mixture is well stirred with a glass rod. It is 
 then let cool and placed in an evaporating dish partly 
 filled with water, in which the melted matter is softened, 
 dissolved out of the crucible into the dish, which is then 
 heated, and the watery solution is filtered into a flask. 
 The residue on the filter, after being well washed, is rinsed 
 back into the dish, very dilute nitric acid is added, and 
 the whole evaporated to dryness. The dry residue is 
 taken up in water acidulated with nitric acid, heated, and 
 filtered into the same flask in which is the aqueous solu- 
 tion. The residue is washed with hot water, the filtrate 
 is allowed to cool in the flask, ferric sulphate or iron 
 alum is added, and the liquid is titrated. 
 
 BLOWPIPE ASSAY OF SILVER. 
 
 The following very complete method for the blowpipe 
 assay of silver and its ores is given by David Forbes, 
 F.K.S., in the 'Chemical News,' JSTos. 380, 384, 392, 396, 
 398, and 412. 
 
 The blowpipe assay of silver ores was first described 
 in 1827 by Harkort,* and subsequently considerably im- 
 proved by Plattner. This assay process is in all cases 
 based upon the reduction to a metallic state of all the 
 silver contained in the compound in question along with 
 more or less metallic lead, which latter metal, when not 
 already present in sufficient quantity in the substance 
 itself under examination, is added in the state of granu- 
 lated lead to the assay previous to its reduction. The 
 globule of silver-lead thus obtained, if soft and free from 
 such elements as would interfere with its treatment upon 
 the cupel, may then be at once cupelled before the blow- 
 pipe until the pure silver alone remains upon the bone-ash 
 surface of the cupel ; but if not, it is previously submitted 
 to a scorifying or oxidising treatment upon charcoal until 
 all such substances are either slagged off or volatilised, 
 and the resulting silver-lead globule cupelled as before. 
 
 * 'Die Probirkunst mit dem Lothrohre.' Freiberg, 1827, I. Heft (all 
 published). 
 
716 
 
 THE ASSAY OF SILVER. 
 
 As, therefore, the final operation in all silver assays is 
 invariably that of cupelling the silver-lead alloy obtained 
 from the previous reduction of the substance, effected by 
 methods differing according to the nature of the argen- 
 tiferous ore or compound under examination, it is here con- 
 sidered advisable to introduce the description of the silver 
 assay by an explanation of this process. 
 
 In the ordinary process of cupellation in the muffle, 
 bone-ash or other cupels are employed of a size large 
 enough to absorb the whole of the litharge produced from 
 the oxidation of the lead in the assay. 
 
 This, however, should not be the case when using the 
 blowpipe ; for as .the heating powers of that instrument 
 are limited, it is found in practice much better to accom- 
 plish this result by two distinct operations the first being 
 a concentration of the silver-lead in which the greater 
 part of the lead is converted by oxidation into litharge re- 
 maining upon, but not, or only very slightly, absorbed by, 
 the bone -ash cupel : and the second in cupelling the small 
 concentrated metallic bead so obtained upon a fresh cupel 
 until the remaining lead is totally absorbed by the cupel 
 and the silver left behind in a pure state. By this means a 
 much larger weight of the silver-lead alloy can be submitted 
 to assay, and, for reasons hereafter to be explained, much 
 more exact results are obtained than 
 would be the case when the cupella- 
 tion is conducted at one operation in 
 the ordinary manner. 
 
 The apparatus used by the author 
 for these operations" are shown to a 
 scale of one-half their real size in the 
 woodcut fig. 140 (a to d). 
 
 In fig. 140, a represents in section a 
 small cylindrical mould of iron, seven- 
 tenths of an inch in diameter, and 
 about four-tenths high, in which is 
 turned a cup-shaped nearly hemisphe- 
 rical depression two-tenths of an inch 
 deep in centre, the inner surface of which is left rough, or 
 
 FIG. 140. 
 
FOKBES'S BLOWPIPE ASSAY. 717 
 
 marked with minute ridges and furrows for the purpose of 
 enabling it to retain more firmly the bone-ash lining which 
 is stamped into it by means of the polished bolt, also shown 
 in the figure. This mould rests upon the stand d, having 
 for this purpose a small central socket in its base, into 
 which the central pivot of the stand enters. This socket 
 is seen in the ground plan, 5, of the base of the mould, 
 which shows likewise three small grooves or slots made to 
 enable a steady hold to be taken of it, when hot, by the 
 forceps. The stand itself is composed of a small turned 
 ivory or wood base, fixed into a short piece of strong glass 
 tubing, which, from its non-conducting powers, serves as an 
 excellent handle. In the centre of the base a slight iron 
 rod rising above the level of the glass outer tube serves as 
 a support for the cupel mould, into the socket in the base 
 of which it enters. 
 
 Bone-ash is best prepared by burning bones which have 
 previously been boiled several .times, so as to extract all 
 animal matter. The best bone-ash is made from the core- 
 bone of the horns of cattle well boiled out and burned. 
 The ash from this is more uniform than from the .other 
 bones, which have in general a very compact enamel-like 
 exterior surface, whilst the interior is of a much softer 
 nature. 
 
 Concentration of the Silver-lead. A cupel is prepared 
 by filling the above-described cupel mould with bone-ash 
 powder not finer than will pass through a sieve containing 
 from forty to fifty holes in the linear inch, and should be 
 well dried and kept in an air-tight bottle, and the whole 
 pressed down with, the bolt, using a few taps of the hammer, 
 It is then heated strongly in the oxidising blowpipe flame,. 
 in order to drive off any hygroscopic moisture. The bone- 
 ash surface of the cupel, after heating, should be smooth, 
 and present no cracks ; if the reverse, these may be re- 
 moved by using the bolt again and reheating.* The silver- 
 lead, beaten on the anvil into the form of a cube, is placed 
 
 * These precautions are very important, as the slightest trace of moisture 
 in the substance of the bone-ash would inevitably cause a spurting of the 
 metal during the operation. 
 
718 THE ASSAY OF SILVER. 
 
 gently upon the surface of the bone-ash, and, directing a 
 pretty strong oxidising flame on to its surface, it is fused, 
 and quickly attains a bright metallic appearance, and 
 commences to oxidise with a rapid rotary movement. 
 (Occasionally, when the assay is large, and much copper or 
 nickel present, the globule may, under this operation, cover 
 itself with a crust of lead oxide or solidify ; in such cases 
 direct the blue point of a strong flame steadily on to one 
 spot of the surface of the lead globule, until it commences 
 oxidising and rotating. In some cases where much nickel 
 is present, an infusible scale, impeding or even preventing 
 this action, may form, but will disappear on adding more 
 lead say from three to six grains, according to the thick- 
 ness of this scale or crust.) When this occurs the cupel 
 is slightly inclined from the lamp, and a fine blue point 
 obtained by placing the blowpipe nozzle deeper into the 
 flame, and the lamp is directed at about an angle of 30 
 on to the globule not, however, so near as to touch it 
 with the blue point, but only with the outer flame, so 
 moderating it as to keep the assay at a gentle red heat, 
 and not allowing the rotation to become too violent. 
 
 This oxidising fusion should be carried on at the lowest 
 temperature sufficient to keep up the rotatory movement, 
 and to prevent a crust of litharge accumulating upon the 
 surface of the globule, but still sufficiently high to hinder 
 the metallic globule from solidifying. Should this, how- 
 ever, happen, a stronger flame must be employed for a 
 moment until the metal is again in rotation ; such inter- 
 ruptions should, however, be avoided. The proper tem- 
 perature can only be learned by practice. A too high 
 temperature is still more injurious, causing the lead to 
 volatilise, and, if rich in silver, carry some of that metal 
 mechanically along with it. The litharge, also, instead of 
 remaining on the cupel, would be absorbed by the bone- 
 ash, and as the surface of the metallic globule is covered 
 by a too thin coating of fused litharge, some silver may be 
 absorbed along with the litharge. In this operation, in 
 order to avoid loss of silver, the fused globule should be 
 always kept in contact with the melted litharge. 
 
FORBES'S BLOWPIPE ASSAY. 719 
 
 By the above treatment, the air has free access to the 
 assay, and the oxidation of the lead and associated foreign 
 metals goes on rapidly. The surface of the melted globule, 
 when poor in silver, shows a brilliant play of iridescent 
 colours, which does not take place when very rich in silver. 
 The litharge is driven to the edge of the globule, heating 
 itself up and solidifying behind and around it. When the 
 globule becomes so hemmed in by the litharge as to present 
 too small a surface for oxidation, the cupel is moved so as 
 to be more horizontal (having been previously kept in an 
 inclined position), thus causing the lead globule to slide 
 by its own weight on one side, and expose a fresh surface 
 to the oxidising action. When the lead is pure, the litharge 
 formed has a reddish-yellow colour, but if copper is 
 present it is nearly black. 
 
 In concentrating silver-lead, it must be remembered 
 that an alloy of lead and silver, if in the proportion of 
 about 86 per cent, silver along with 14 per cent, lead, 
 when cooled slowly in the litharge behaves in a manner 
 analogous to the spitting of pure silver, throwing out a 
 whitish-grey pulverulent excrescence rich in silver. For 
 this reason, therefore, the concentration above described 
 should be stopped when the globule is supposed to contain 
 about six parts silver along with one part in weight of 
 lead. In case, however, this limit should have been ex- 
 ceeded, it is advisable at once to push the concentration 
 still further until the silver globule contains but very little 
 lead. In practice with poor ores it is usual to concentrate 
 the lead until the globule is reduced to the size of a small 
 mustard -seed, or in rich ores to some two or three times 
 that size. Upon arriving at this point, the cupel is with- 
 drawn very gradually from the flame, so that the cooling 
 shall take place as slowly as possible until the globule has 
 solidified in its envelope of litharge. If cooled too quickly, 
 the litharge, contracting suddenly, would throw out the 
 globule, or even cause it to spirt; in such case it should 
 be touched by the point of the blue flame, so as to fuse it 
 to a round globule, which is cooled slowly, as before de- 
 scribed. The globule is now reserved for the next opera- 
 
720 THE ASSAY OF SILVER. 
 
 tion, for which purpose it is, when quite cold, extracted 
 from the litharge surrounding it. 
 
 Cupellation.- The bone- ash required for this process 
 should be of the best quality and in the most impalpable 
 powder, prepared by elutriating finely ground bone ash, 
 and drying the product before use. 
 
 The cupel, still hot from the last operation, is placed 
 upon the anvil, and the crust of litharge, with its enclosed 
 metallic bead, gently removed, leaving the hot coarse bone- 
 ash beneath it in the mould ; upon this a small quantity 
 of the elutriated bone-ash is placed, so as to fill up the 
 cavity, and the whole, whilst hot, stamped down by the bolt, 
 previously slightly warmed, with a few taps of the ham- 
 mer. The cupel thus formed is heated strongly in the 
 oxidising flame, which should leave the surface perfectly 
 smooth, and free from any fissures or scales ; if such 
 appear, the bolt must again be used, and the cupel re- 
 heated. In this process it is very important that the cupel 
 should possess as smooth a surface as possible, whilst at 
 the same time the substance of the cupel beneath should 
 not be too compact, so as thereby to permit the litharge 
 to filter through and be readily absorbed, leaving the silver 
 bead upon the smooth upper surface. 
 
 The bead of silver-lead obtained from the last operation 
 is taken out of the litharge in which it is embedded, and, 
 after removing any trace of adherent bone-ash or litharge, 
 is slightly flattened to prevent its rolling about upon the 
 surface of the cupel. 
 
 It is now put into the cupel prepared as before de- 
 scribed, placing it on the side farthest from the lamp and 
 a little above the centre of the cupel, which is now in- 
 clined slightly towards the lamp, and is heated by the 
 oxidising flame directed downwards upon it, this causing 
 the globule, when fused and oxidising, to move of itself 
 into the centre of the cupel. The cupel is now brought 
 into a horizontal position, and the flame, directed on to it at 
 an angle of about forty-five degrees, is made to play upon 
 the bone-ash surface immediately surrounding the globule, 
 without, however, touching it, so as to keep this part of 
 
FORBESS BLOWPIPE ASSAY. 721 
 
 the cupel at a red heat sufficiently strong to insure the 
 globule being in constant oxidising fusion, and at the same 
 time to cause the perfect absorption of the litharge, so as 
 to prevent any scales of litharge forming upon the surface 
 of the cupel under the globule, which would impede the 
 oxidation, as well as prevent the silver bead being easily 
 detached at the conclusion of the operation. Should the 
 heat at any time be too low and the globule solidify, it 
 must be touched for an instant with the point of the flame 
 and proceeded with as before. Should (in consequence of 
 the bone-ash not having been sufficiently heated to absorb 
 the litharge perfectly) a little litharge adhere pertinaciously 
 to the globule, or a particle of the bone-ash cupel attach 
 itself, the cupel should be slightly inclined, so as to allow 
 the globule to move by its own weight on to another and 
 clean part of the cupel, leaving the litharge or bone-ash 
 behind it ; but, if not sufficiently heavy to do so, a small 
 piece of pure lead may be fused to it in order to increase its 
 weight, and so allow of the same proceeding being adopted. 
 
 By slightly inclining the cupel-stand, and moving it so 
 as to present in turn all parts of the surface surrounding 
 the globule to the action of the flame, the cupellation pro- 
 ceeds rapidly. If, however, the assay contains very little 
 silver, it will be found necessary to move the globule from 
 one spot to another on the cupel, in order to present a fresh 
 surface for absorbing the litharge formed ; this is done by 
 simply inclining the cupel-stand, remembering that the bone- 
 ash surrounding the globule must always be kept at a red 
 heat, without ever touching the globule itself by the flame. 
 
 In assays rich in silver a play of iridescent colours 
 appears some seconds before the ' brightening,' which dis- 
 appears the moment the silver becomes pure ; as soon as 
 this is observed the cupel should be moved in a circular 
 manner, so that the globule is nearly touched all round by 
 the point of the blue flame, and this is continued until the 
 surface of the melted silver is seen to be quite free from 
 any litharge, upon which it is very gradually withdrawn 
 from the flame so as to cool the assay by degrees very 
 slowly, in order to prevent ' spitting.' 
 
 3 A 
 
722 THE ASSAY OF SILVER. 
 
 When the silver-lead is very poor, this play of colours 
 is not apparent, and as soon as the rotatory movement of 
 the globule ceases, the heat must be increased for an 
 instant, in order to remove the last thin but pertinacious 
 film or scale of litharge, and subsequently the assay is 
 cooled gradually ; when cold it should, whilst still upon 
 the cupel, be examined by a lens, to see whether the bead 
 possesses a pure silver colour, as, if not, it must be reheated. 
 
 Frequently, when the ' brightening ' takes place, the 
 silver globule is found to spread out, and, after cooling, 
 although of a white colour, is found to appear somewhat 
 less spherical or more flattened in shape than a correspond- 
 ing globule of pure silver would be. This arises from the 
 presence of copper still remaining in the silver, and in such 
 cases a small piece of pure lead (about from one- half to 
 one and a half grain in weight, according to size of assay) 
 should be fused on the cupel along with the silver, and the 
 cupellation of the whole conducted as before on another 
 part of the cupel, when the silver globule will be obtained 
 pure, and nearly spherical in shape. Sometimes the silver 
 globule in ' brightening ' may still remain covered with a 
 thin film of litharge, although otherwise pure ; this arises 
 from too little heat having been employed in the last stage 
 of the operation, and consequently the bead should be 
 re-heated in a strong oxidising flame until this litharge is 
 absorbed, and the globule, after slow cooling, appears pure. 
 
 If the instructions here given be strictly attended to, it 
 will be found after some practice that very accurate results 
 may be obtained in the blowpipe assay for silver, and that 
 no difficulty will be found in detecting the presence and 
 estimating the amount of silver present, even when in as 
 small a quantity as half an ounce to the ton. When sub- 
 stances containing very little silver or less than that amount 
 are examined, several assays should be made, and the 
 silver-lead obtained concentrated separately, after which 
 the various globules should be united and cupelled together 
 in one operation. 
 
 It is hardly necessary to remark that the lead employed 
 in assaying should be free from silver, or, if not, its actual 
 
FORBES'S BLOWPIPE ASSAY. 723 
 
 contents in silver should be estimated, and subtracted from 
 the amount found in the assay. 
 
 Assay lead containing less than one quarter of an ounce 
 to the ton of lead can readily be obtained, or can be made 
 by precipitating a solution of acetate of lead by metallic 
 zinc, rejecting the first portion of lead thrown down. In 
 all cases the lead should be fused and granulated finely 
 the granulated lead for use in these assays being previously 
 passed through a sieve containing forty holes to the linear 
 inch. It is also useful to have some lead in the form of 
 wire, as being very convenient for adding in small portions 
 to assays when on the cupel. 
 
 Estimation of the Weight of the Silver Globule obtained 
 on Cupellation. As the amount of lead which can, by 
 the method before described, be conveniently cupelled 
 before the blowpipe is necessarily limited, the silver glo- 
 bule which remains upon the bone-ash surface of the cupel 
 at the end of the operation is, when substances poor in 
 silver have been examined, frequently so very minute that 
 its weight could not be estimated with correctness by 
 the most delicate balances in general use. 
 
 The blowpipe balance employed by the author turns 
 readily with one-thousandth of a grain, but could not be 
 used for estimating weights below that amount. 
 
 Globules of silver of far less weight than one-thousandth 
 are distinctly visible to the naked eye a circumstance 
 which induced Harkort to invent a volumeti|ical scale 
 based upon the measurement of the diameters of the glo- 
 bules, which scale in practice has been found of very great 
 utility in the blowpipe assay of silver. 
 
 The scale for this purpose which is employed by the 
 author is shown in full size in the woodcut on p;. 724. 
 
 This figure represents a small strip of highly polished 
 ivory about 6J inches long, f inch broad, and inch in 
 thickness, on which are drawn, by an extremely fine point, 
 two very fine and distinct lines emanating from the lower 
 or zero point, and diverging upwards until, at the distance 
 of exactly six English standard inches, they are precisely 
 four hundredth parts of an inch apart. This distance 
 
 3 A 2 
 
724 
 
 THE ASSAY OF SILVER. 
 
 (six inches) is, as shown in woodcut, divided into 100 equal 
 parts by cross lines numbered in accordance from zero 
 FIG. i4i. upwards. It is now evident if a small globule 
 of silver be placed in the space between these 
 two lines, using a magnifying glass to assist 
 the eye in moving it up or down until the 
 diameter of the globule is exactly contained 
 within the lines themselves, that we have at 
 once a means of estimating the diameter of 
 the globule itself, and therefrom are enabled 
 to calculate its weight. 
 
 As the silver globules which cool upon 
 the surface of the bone-ash cupel are not true 
 spheres, but are considerably flattened on the 
 lower surface, where they touch and rest upon 
 the cupel, it follows that the weight of glo- 
 bules corresponding in diameter to the extent 
 of divergence at the different degrees of the 
 scale cannot be calculated directly from their 
 diameters as spheres, but require to have 
 their actual weight experimentally determined 
 in the same manner as employed byPlattner. 
 .. -The table here appended has been cal- 
 culated by the author, and in one column 
 shows the diameter in English inches corre- 
 sponding to each number or degree of the 
 scale itself, and in the two next columns the 
 .respective weights of the flattened spheres 
 which correspond to each degree or diameter ; 
 for convenience these weights are given in 
 the different columns in decimals, both of 
 English grains and of French grammes. 
 
 These weights are calculated from the 
 following data, found as the average result of 
 several very careful and closely approximating 
 assays, which showed that globules of silver 
 exactly corresponding to No. 95 on this scale, 
 or 0*038 inch in diameter, possessed a weight 
 of 0-0475573 grains or 0-003079 grammes. Erom this the 
 
 100 
 98 
 
 
 
 96 
 94 
 
 
 
 92 _: 
 
 90 : 
 
 88 " 
 
 1- : 
 
 86 
 
 
 84 
 
 
 82 
 
 
 80 ~ 
 
 
 78 
 
 
 76 ~ 
 
 
 74 
 
 
 72 
 70 
 68 - 
 66 -f 
 64 
 62 - 
 60 
 68 - 
 56 7 
 
 :~- 
 
 
 
 50 : 
 
 48 - 
 46 
 
 44 
 v 
 42 - 
 
 40 - 
 38 
 36 - 
 '34 
 ; 32 - 
 
 5 
 
 28-^ 
 
 
 W 
 
 24 
 22 
 SO 
 18 -^ 
 16 -3 
 14 
 12 
 
 | 
 
 i 10 
 
 
 6 ~ 
 
 
 
 
 o * 
 
 
FORBES 8 BLOWPIPE ASSAY. 
 
 725 
 
 No. on 
 scale 
 
 Greatest diameter 
 in inches 
 
 Weight of globule in 
 grains 
 
 Weight of globule in 
 grajnmes 
 
 1 
 
 0-0004 
 
 0-00000005 
 
 0-000000003 
 
 2 
 
 0-0008 
 
 0-00000044 
 
 0-000000028 
 
 3 
 
 0-0012 
 
 0-00000149 
 
 0-000000096 
 
 4 
 
 0-0016 
 
 0-00000355 
 
 0-000000229 
 
 5 
 
 0-0020 
 
 0-0000069 
 
 0-00000044 
 
 6 
 
 0-0024 
 
 0-0000119 
 
 0-00000077 
 
 7 
 
 0-0028 
 
 0-0000190 
 
 0-00000120 
 
 8 
 
 0-0032 
 
 0-0000284 
 
 0-00000184 
 
 9 
 
 0-0036 
 
 0-0000403 
 
 0-00000262 
 
 10 
 
 0-0040 
 
 0-0000554 
 
 0-00000359 
 
 11 
 
 0-0044 
 
 0-0000736 
 
 0.00000478 
 
 12 
 
 0-0048 
 
 0-0000958 
 
 0-00000620 
 
 13 
 
 0-0052 
 
 0-0001218 
 
 0-00000789 
 
 14 
 
 0-0056 
 
 0-0001522 
 
 0-00000985 
 
 15 
 
 0-0060 
 
 0-0001872 
 
 0-00001203 
 
 16 
 
 0-0064 
 
 0-0002272 
 
 0-00001471 
 
 17 
 
 0-0068 
 
 0-0002725 
 
 0-00001764 
 
 18 
 
 0-0072 
 
 0.0003234 
 
 0-00002094 
 
 19 
 
 0-0076 
 
 0-0003804 
 
 0-00002463 
 
 20 
 
 0-0080 
 
 0-0004437 
 
 0-00002872 
 
 21 
 
 0-0084 
 
 0-0005137 
 
 0-00003327 
 
 22 
 
 0-0088 
 
 0-0005906 
 
 0-00003823 
 
 23 
 
 0-0092 
 
 0-0006748 
 
 0-00004367 
 
 24 
 
 0-0096 
 
 0-0007668 
 
 0-00004964 
 
 25 
 
 0-0100 
 
 0-0008667 
 
 0-00005611 
 
 26 
 
 0-0104 
 
 0-0009749 
 
 0-00006311 
 
 27 
 
 0-0108 
 
 0-0010918 
 
 0-00007068 
 
 28 
 
 0-0112 
 
 0-0012176 
 
 0-00007883 
 
 29 
 
 0-0116 
 
 0-0013528 
 
 0-00008758 
 
 30 
 
 0-0120 
 
 0-0014976 
 
 0-00009696 
 
 31 
 
 0-0124 
 
 0-0016524 
 
 0-00010698 
 
 32 
 
 0-0128 
 
 0-0018176 
 
 0-00011677 
 
 33 
 
 0-0132 
 
 0-0019934 
 
 0-00012817 
 
 34 
 
 0-0136 
 
 0-0021801 
 
 0-00014114 
 
 35 
 
 0-0140 
 
 0-0023786 
 
 0-00015397 
 
 36 
 
 0-0144 
 
 0-0025879 
 
 0-00016755 
 
 37 
 
 0-0148 
 
 0-0028097 
 
 0-00018190 
 
 38 
 
 0-0152 
 
 0-0030437 
 
 0-00019705 
 
 39 
 
 0-0156 
 
 0-0032903 
 
 0-00021302 
 
 40 
 
 0-0160 
 
 0-0035550 
 
 0-00022983 
 
 41 
 
 0-0164 
 
 0-0038230 
 
 0-00024751 
 
 42 
 
 0-0168 
 
 0-0041096 
 
 0-00026606 
 
 43 
 
 0-0172 
 
 0-0044111 
 
 0-00028553 
 
 44 
 
 0-0176 
 
 0-0047250 
 
 0-00030589 
 
 45 
 
 G'0180 
 
 0-0050546 
 
 0.00032725 
 
 46 
 
 0-0184 
 
 0-0053991 
 
 0-00034955 
 
 47 
 
 0-0188 
 
 0-0057590 
 
 0-00037285 
 
 48 
 
 0-0192 
 
 0-0061344 
 
 0-00039716 
 
 49 
 
 0-0196 
 
 0-0065258 
 
 0-00042250 
 
 50 
 
 0-0200 
 
 0-0069335 
 
 0-00044890 
 
 51 
 
 0-0204 
 
 0-0073581 
 
 0-00047638 
 
 52 
 
 0-0208 
 
 0-0077799 
 
 0-00050495 
 
 53 
 
 0-0212 
 
 0-0082580 
 
 0-00053464 
 
 54 
 
 0-0216 
 
 0-00873438 
 
 0-00056549 
 
 55 
 
 0-0220 
 
 0-00922854 
 
 0-00059748 
 
 56 
 
 0-0224 
 
 0-0097412 
 
 0-00063067 
 
726 
 
 THE ASSAY OF SILVEK. 
 
 Jfo. on 
 scale 
 
 Greatest diameter 
 in inches 
 
 Weight of globule in 
 grains 
 
 Weight of globule in 
 grammes 
 
 57 
 
 0-0228 
 
 0-0102725 
 
 0-00066506 
 
 58 0-0232 
 
 0-0108228 
 
 0-00070021 
 
 59 0-0236 
 
 0-0113922 
 
 0-00073753 
 
 60 0-0240 
 
 0-0119815 
 
 0-00077570 
 
 61 0-0244 
 
 0-0125901 
 
 0-00081513 
 
 62 0-0248 
 
 0-0132119 
 
 0-00085588 
 
 63 0-0252 
 
 0-0138901 
 
 0-00089797 
 
 64 0-0256 
 
 0-0145440 
 
 0-00094141 
 
 65 0-0260 
 
 0-0152311 
 
 0-00098623 
 
 66 
 
 0-0264 
 
 0-0159472 
 
 0-00103245 
 
 67 
 
 0-0268 
 
 0-0166828 
 
 0-00108010 
 
 68 0-0272 
 
 0-0174414 
 
 0-00112918 
 
 69 
 
 0-0276 
 
 0-0182220 
 
 0-00117974 
 
 70 
 
 0-0280 
 
 0-0190256 
 
 0-00123177 
 
 71 
 
 0-0284 
 
 0-0198529 
 
 0-00128535 
 
 72 
 
 0-0288 
 
 0-0207035 
 
 0-00134041 
 
 73 
 
 0-0292 
 
 0-0215782 
 
 0-00139704 
 
 74 
 
 0-0296 
 
 0-0224469 
 
 0-00145525 
 
 75 0-0300 
 
 0-0234010 
 
 0-00151504 
 
 76 0-0304 
 
 0-0243496 
 
 0-00157645 
 
 77 
 
 0-0308 
 
 0-0253224 
 
 0-00163950 
 
 78 
 
 0-0312 
 
 0-0263228 
 
 0-00170422 
 
 79 
 
 0-0316 
 
 0-0273484 
 
 0-00177060 
 
 80 
 
 0-0320 
 
 0-0284000 
 
 0-00183869 
 
 81 
 
 0-0324 
 
 0-0294789 
 
 0-00190852 
 
 82 
 
 0-0328 
 
 0-0305838 
 
 0-00198008 
 
 83 
 
 0-0332 
 
 0-0317162 
 
 0-00205340 
 
 84 
 
 0-0336 
 
 0-0328768 
 
 0-00212851 
 
 85 
 
 0-0340 
 
 0-0340649 
 
 0-00220549 
 
 86 
 
 0-0344 
 
 0-0349739 
 
 0-00228400 
 
 87 
 
 0-0348 
 
 0-0364422 
 
 0-00235938 
 
 88 
 
 0-0352 0-0378008 
 
 0-00244730 
 
 89 
 
 0-0356 0-0390138 
 
 0-00253168 
 
 90 
 
 0-0360 0-0404368 
 
 0-00261797 
 
 91 
 
 0-0364 0-0417943 
 
 0-00270790 
 
 92 
 
 0-0368 0-0431930 
 
 0-00279642 
 
 93 
 
 0-0372 0-0446162 
 
 0-00288860 
 
 94 
 
 0-0376 0-0460718 
 
 0-00298279 
 
 95 
 
 0-0380 0-0475573 
 
 0-00307900 
 
 96 
 
 0-0384 0-0465239 
 
 0-00317728 
 
 97 
 
 0-0388 0-0506249 
 
 0-00327759 
 
 98 
 
 0-0392 0-0522069 
 
 0-00338020 
 
 99 
 
 0-0396 0-0538215 
 
 0-00348452 
 
 100 
 
 0-0400 0-0554688 
 
 0-00359138 
 
 respective weights of all the other numbers or degrees on 
 this scale were calculated, on the principle that solids were 
 to one another in the ratio of the cubes of their diameters. 
 This mode of calculation is not, however, absolutely correct 
 in principle, for the amount of flattening of the under sur- 
 face of the globule diminishes in reality with the decreasing 
 
FORBES'S BLOWPIPE ASSAY. 727 
 
 volume of the globule. In actual practice, however, this 
 difference may be assumed to be so small that it may be 
 neglected without injury to the correctness of the results. 
 
 The smaller the diameter of the globule, the less will 
 be the difference or variation in weight in descending the 
 degrees of this scale, since the globules themselves vary in 
 weight with the cubes of their diameters : for this reason, 
 also, all such globules as come within the scope of the 
 balance employed should be weighed in preference to being 
 measured, and this scale should be regarded as more spe- 
 cially applicable to the smaller globules beyond the reach 
 of the balance. 
 
 Cupellation Loss. This term is applied to indicate a 
 minute loss of silver, unavoidably sustained in the process 
 of cupellation, which arises from a small portion of that 
 metal being mechanically carried along with the litharge 
 into the body of the cupel. The amount of this loss in- 
 creases with the quantity of lead present in the assay 
 (whether contained originally in the assay or added subse- 
 quently for the purpose of slagging off the copper, &c.) ; 
 it is relatively greater as the silver globule is larger, but 
 represents a larger percentage of the silver actually con- 
 tained in the assay, in proportion as the silver globule 
 obtained diminishes in size. It has, however, been experi- 
 mentally proved that, in assays of like richness in silver, 
 this loss remains constant when the same temperature has 
 been employed, and similar weights of lead have been 
 oxidised in the operation. 
 
 In the blowpipe assay this loss is not confined to the 
 ultimate operation of cupellation, but occurs, though in a 
 less degree, in the concentration of the silver-lead, and in 
 the previous scorification of the assay, had such operation 
 preceded the concentration. The total loss in the blow- 
 pipe assay is found, however, to be less than in the ordinary 
 muffle assay, since in the latter case the whole of the oxi- 
 dised lead is directly absorbed by the cupel. 
 
 In mercantile assays of ore it is not customary to pay 
 attention to the cupellation loss, and the results are usually 
 stated in the weight of silver actually obtained. Where, 
 
728 
 
 THE ASSAY OP SILVER. 
 
 however, great accuracy is required, especially when the 
 substances are very rich in silver, the cupellation loss is 
 added to the weight of the silver globule obtained, in order 
 to arrive at the true percentage. 
 
 The amount to be added for this purpose is shown 
 in the annexed table, which is slightly modified from 
 Plattner's : 
 
 Actual per- 
 centage of 
 silver found by 
 assay 
 
 Cupellation loss, or percentage of silver to be added to the actual percentage 
 found by assay in order to show the true percentage of silver contained in same. 
 The entire amount of lead in or added to the assay being the folio wing multiples 
 of the original weight of assay : 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 I 8 
 
 1 
 
 11 
 
 13 
 
 16 
 
 1 
 
 99-5 / 
 
 0-25 
 
 0-32 
 
 0-39 
 
 0-45 
 
 0-50 
 
 
 
 
 
 
 90 
 
 0-22 
 
 0-29 
 
 0-36 
 
 0-42 
 
 0-47 
 
 0-69 
 
 0-83 
 
 
 
 
 80 
 
 0-20 
 
 0-26 
 
 0-33 
 
 0-39 
 
 0-44 
 
 0-64 
 
 0-75 
 
 
 
 
 70 
 
 0-18 
 
 0-23 
 
 0-29 
 
 0-35 
 
 0-40 
 
 0-58 
 
 0-68 
 
 0-82 
 
 
 
 60 
 
 0-16 
 
 0-20 
 
 0-26 
 
 0-30 
 
 0-36 
 
 0-52 
 
 0-61 
 
 0-74 
 
 
 
 50 
 
 0-14 
 
 0-17 
 
 0-23 
 
 0-26 
 
 0-32 
 
 0-46 
 
 0-54 
 
 0-65 
 
 
 
 40 
 
 0-12 
 
 0-15 
 
 0-20 
 
 0-22 
 
 0-27 
 
 0-39 
 
 0-46 
 
 0-55 
 
 0-62 
 
 
 35 
 
 0-11 
 
 0-13 
 
 0-18 
 
 0-18 
 
 0-25 
 
 0-36 
 
 0-42 
 
 0-50 
 
 0-57 
 
 
 30 
 
 0-10 
 
 0-12 
 
 0-16 
 
 0-16 
 
 0-22 
 
 0-32 
 
 0-38 
 
 0-45 
 
 0-51 
 
 
 25 
 
 0-09 
 
 0-10 
 
 0-14 
 
 0-14 
 
 0-20 
 
 0-29 
 
 0-34 
 
 0-40 
 
 0-45 
 
 
 20 
 
 0-08 
 
 0-09 
 
 0-12 
 
 0-12 
 
 0-17 
 
 0-25 
 
 0-29 
 
 0-35 
 
 0-39 
 
 0-45 
 
 15 
 
 0-07 
 
 0-08 
 
 0-10 
 
 0-11 
 
 0-15 
 
 0-20 
 
 0-23 
 
 0-28 
 
 0-32 
 
 0-37 
 
 12 
 
 0-06 
 
 0-07 
 
 0-09 
 
 0-10 
 
 0-13 
 
 0-17 
 
 0-19 
 
 0-23 
 
 0-26 
 
 0-22 
 
 10 
 
 0-05 
 
 0-06 
 
 0-08 
 
 0-09 
 
 0-11 
 
 0-15 i 0-17 
 
 0-20 
 
 0-23 
 
 0-27 
 
 9 
 
 0-04 
 
 0-05 
 
 0-07 
 
 0-08 
 
 0-10 
 
 0-14 
 
 0-16 
 
 0-18 
 
 0-21 
 
 0-25 
 
 8 
 
 0-03 
 
 0-04 
 
 0-06 
 
 0-07 
 
 0-09 
 
 0-13 
 
 0-15 
 
 0-16 
 
 0-18 
 
 0-22 
 
 7 
 
 0-02 
 
 0-03 
 
 0-05 
 
 0-06 
 
 0-08 
 
 0-12 i 0-13 
 
 0-14 
 
 0-16 
 
 0-20 
 
 6 
 
 0-01 
 
 0-02 
 
 0-04 
 
 0-05 
 
 0-07 
 
 0-10 
 
 0-11 
 
 0-12 
 
 0-14 
 
 0-17 
 
 5 
 
 
 0-91 
 
 0-03 
 
 0-04 
 
 0-06 
 
 0-09 
 
 0-10 
 
 0-11 
 
 0-12 
 
 0-14 
 
 4 
 
 
 
 0-02 
 
 0-03 
 
 0-05 i 0-07 
 
 0-08 
 
 0-09 
 
 0-10 
 
 0-11 
 
 3 
 
 
 
 0-01 
 
 0-02 
 
 0-04 
 
 0-05 
 
 0-06 
 
 0-07 
 
 0-08 
 
 0-09 
 
 2 
 
 
 
 
 0-01 
 
 0-03 
 
 0-04 
 
 0-04 
 
 0-05 
 
 0-06 
 
 0-07 
 
 1 
 
 
 
 
 , 
 
 0-01 
 
 0-03 
 
 0-03 
 
 0-04 
 
 0-04 
 
 0-05 
 
 The use of this table is best explained by an example, 
 as the following : An assay to which there had been 
 added, in all, five times its weight of assay lead, gave a 
 globule of silver equivalent to six per cent. Upon referring 
 to the table it will be seen that the cupellation loss for this 
 would be 0*07 ; consequently the true percentage of silver 
 contained in the assay would be 6'07. This table is only 
 extended to whole numbers, but fractional parts can easily 
 be calculated from the same. 
 
FOEBES'S BLOWPIPE ASSAY. 729 
 
 When the globules of silver are so minute that they 
 cannot be weighed, but must be measured upon the scale, 
 the cupellation loss should not be added, since, as a rule, 
 it would be less than the difference which might arise from 
 errors of observation likely to occur when measuring their 
 diameters upon the scale. 
 
 In the case of beginners, it will be found that the cupel- 
 lation is usually carried on at too high a temperature, and 
 that thereby a greater loss is occasioned than would be 
 accounted for by the above table. After some trials the 
 necessary experience will be acquired in keeping up the 
 proper temperature at which this operation should be 
 effected. 
 
 It now becomes necessary to consider in detail the 
 processes requisite for extracting the silver contents (in 
 combination with lead) from the various metallic alloys of 
 silver which are met with in nature or produced in the 
 arts. 
 
 In considering these, the following classification of the 
 substances will be found convenient : 
 
 METALLIC ALLOYS. 
 
 A. Capable of direct Cupellation. 
 
 a. Consisting chiefly of lead or bismuth ; silver-lead and 
 argentiferous bismuth, native bismuthic silver. 
 
 6. Consisting chiefly of silver : native silver, bar silver, test 
 silver, precipitated silver, retorted silver amalgam, stan- 
 dard silver, alloys of silver with gold and copper. 
 
 c. Consisting chiefly of copper : native copper, copper ingot, 
 sheet or wire, cement copper, copper corns, copper- 
 nickel alloys. 
 
 B. Incapable of direct Cupellation. 
 
 a. Containing much copper or nickel, with more or less 
 sulphur, arsenic, zinc, &c. ; unrefined or black copper, 
 brass, German silver. 
 
 6. Containing tin ; argentiferous tin, bronze, bell-metal, gun- 
 metal, bronze coinage. 
 
 c. Containing antimony, tellurium, or zinc. 
 
 d. Containing mercury : amalgams. 
 
 e. Containing much iron : argentiferous steel, bears from 
 
 smelting furnaces. 
 
730 THE ASSAY OF SILVER. 
 
 A. METALLIC ALLOYS CAPABLE OF DIBECT CUPELLATION. 
 
 a. Consisting chiefly of Lead or Bismuth. In estimat- 
 ing the silver contained in these alloys, it is only requisite 
 to place a clean piece of the same, weighing about from 
 one to ten grains according to its probable richness in 
 silver, upon a cupel of coarse bone-ash, and proceed by 
 concentration and cupellation exactly as has been already 
 described under these heads. 
 
 Should the substance be not altogether metallic, or 
 not free from adherent slag, earthy or other extraneous 
 matter, it should previously be fused on charcoal with a 
 little borax in the reducing blowpipe flame, and the clean 
 metallic globule then removed from the charcoal, and 
 treated as before. In order to remove the globule from 
 the adherent borax-glass, it may be allowed to cool, and 
 then detached ; or, after a little practice, it will be found 
 easy, by a quick movement of the charcoal, to cause the 
 globule, still melted, to detach itself completely, and drop 
 on the anvil in the form of a single somewhat flattened 
 globule, without suffering any loss of lead adhering to the 
 charcoal. 
 
 In the case of argentiferous bismuth alloys the process 
 is carried on in all respects the same as if silver-lead were 
 being treated. As, however, the bismuth globule is very 
 brittle, care must be taken when separating the concen- 
 trated globule from the litharge, as, if not carefully done, 
 a loss may easily be sustained from a portion of the glo- 
 bule remaining behind adherent to the litharge. It is 
 better, therefore, to remove the litharge by degrees from 
 the globule with the aid of the forceps. 
 
 Argentiferous bismuth, free from lead, when cupelled 
 alone, invariably leaves a globule of silver, having a dull 
 frosted surface. If, however, at the end of the operation 
 a small quantity of lead 4 (i to ^ a grain) be added, and 
 fused along with it, the silver globule then obtained will 
 be perfectly bright and free from all bismuth. 
 
 In the case of native bismuthic silver it is advisable to 
 
FOKBES'S BLOWPIPE ASSAY. 731 
 
 fuse the previously weighed mineral with a little lead and 
 borax-glass on charcoal in the reducing flame, so as to 
 free it from any adherent earthy matter, and then proceed 
 by concentration and cupellation, as before described. 
 
 b. Consisting chiefly of Silver: native silver, bar, test, 
 and precipitated silver, retorted silver amalgam, standard 
 silver, silver coin and other alloys of silver with gold and 
 copper. These alloys may be at once fused with lead 
 on the cupel itself, and the operation finished as before 
 described. In general, however, it is better to fuse the 
 weighed assay previously with the requisite amount of 
 pure lead and a little borax-glass, say from a quarter to 
 half the weight of assay, in the reducing flame at a low 
 heat on charcoal until the globule commences to rotate. 
 This insures having a perfectly clean button of silver-lead, 
 which is then cupelled in the ordinary manner. 
 
 In most cases the quantity of lead to be added need 
 not exceed that of the weight of the alloy, but when 
 several percentages of copper are present in the assay, as 
 in the case of many coins, &c., the lead should be increased 
 to some three, or even five, times the weight of the assay 
 in proportion to the amount of copper actually contained 
 in the substance under examination, and which will be 
 treated of more at length under the head of copper-silver 
 alloys. 
 
 When no more lead has been added to the assay than 
 its own weight, the cupellation may be concluded in one 
 operation by inclining the stand, and so moving the glo- 
 bule to a clean part of the cupel ; but when more copper 
 is present, it is preferable to concentrate first and cupel 
 subsequently, in order thereby to reduce the cupellation 
 loss to its minimum. 
 
 In the concentration as much copper as possible should 
 be slagged off with the lead, which is effected by inclining 
 the cupel somewhat more than usual, so that its surface 
 may be less covered up with the litharge and exposed as 
 much as possible to oxidation, by which means the litharge, 
 as it forms, is enabled to carry off more of the copper 
 contained in the silver-lead. 
 
732 THE ASSAY OF SILVER. 
 
 Should the silver globule after cupellation show indi- 
 cations of still containing copper, as before noticed, when 
 treating of cupellation, a small quantity of lead must be 
 fused along with it, and the cupellation finished as usual. 
 
 As at the present time no means are known by which 
 silver can be separated from gold by the use of the blow- 
 pipe in all cases of alloys containing gold, this metal 
 remains to the last along with the silver, and the result in 
 such cases always indicates the combined weight of both 
 these metals contained in the alloy under examination. 
 The wet assay must be resorted to for effecting their 
 separation. 
 
 c. Containing chiefly Copper : native copper, ingot, wire 
 or sheet copper, cement copper, copper coins, copper-nickel 
 alloys. Under the most favourable conditions in cupella- 
 tion, the amount of lead requisite, when converted into 
 litharge, to slag off one part of copper along with it as 
 oxide, amounts to between seventeen and eighteen times 
 its weight. In the blowpipe assay it is usual to add to 
 any cupriferous alloy an amount of pure lead equal to 
 twenty times the amount of copper contained in the alloy, 
 in order to insure the whole of the copper being separated 
 in the litharge. In the case of nickel the amount of lead 
 required is somewhat less than with copper, but in prac- 
 tice the same amount of lead may be employed. 
 
 When the copper is quite clean the requisite amount 
 of lead may be added to it in a single piece on the cupel, 
 fused and cupelled as usual, after previous concentration 
 of the silver-lead to a small-sized globule. 
 
 It is generally found, however, that traces of iron, slag, 
 gangue, or other foreign matter, are present ; and, conse- 
 quently, it is usually advisable to fuse the assay along 
 with the requisite amount of lead, and about one-half its 
 own weight of borax-glass in the reducing flame, until the 
 whole of the substance is seen to have perfectly combined 
 or alloyed with the lead, and the globule has entered into 
 brisk rotation, whilst at the same time no detached 
 metallic globules are seen in the borax-glass. 
 
 The concentration of the silver-lead and cupellation 
 
FORBES'S BLOWPIPE ASSAY. 733 
 
 are then conducted as usual, taking care when concentrat- 
 ing to incline the cupel-stand so as to expose as much as 
 possible of the metallic surface of the melted globule to 
 the oxidising action of the air, with a view of enabling the 
 litharge whilst forming to carry off as much copper along 
 with it as possible. 
 
 Should the silver globule obtained after cupellation 
 spread out, or appear to the eye more flattened than 
 usual with globules of pure silver, it indicates that some 
 copper still remains, and a small piece of assay lead (-J to 
 1 grain weight) should be placed alongside it whilst still 
 on the cupel, fused together, and the cupellation finished 
 on a clean part of the same cupel as usual. 
 
 Precipitated or cement copper, especially that which 
 is in the crude state, and has not been melted and run 
 into ingots, is often very impure, containing so much iron, 
 lead, arsenic, earthy matter, &c., as not to admit of direct 
 cupellation, and in such case should be treated as pertain- 
 ing to class B. a. 
 
 B. METALLIC ALLOYS INCAPABLE OF DIRECT CUPELLATION. 
 
 a. Containing much Copper or Nickel, with frequently 
 some little sulphur, arsenic, zinc, iron, cobalt, fyc. ; as unre- 
 fined or black copper, brass, German silver, fyc. As the 
 presence of these extraneous matters would interfere with 
 the cupellation, either by causing a loss of silver-lead 
 projected from the cupel upon the evolution of the 
 volatile substances present, or by forming oxides which 
 could not be absorbed by the cupel, it is necessary to 
 eliminate such substances by a scorification with borax 
 on charcoal, previous to concentration or cupellation. 
 
 In the case of unrefined and black copper, the portion 
 used in the examination is placed in the scoop with twenty 
 times its weight of assay lead, and its own weight of pow- 
 dered borax-glass, mixed with the spatula, and transferred 
 to a soda-paper cornet. It is then fused on charcoal in 
 the reducing flame, which should be constant and uninter- 
 rupted, until all particles have completely united, and a 
 
734 THE ASSAY OF SILVER. 
 
 brisk rotation sets in, which is kept up for a short time, 
 when the silver-lead globule, which should appear bright 
 on the surface after cooling, is concentrated and cupelled 
 precisely as is directed under A. c. By this preliminary 
 scorification the sulphur, arsenic, and zinc are volatilised, 
 and any lead, cobalt, or iron slagged off into the borax- 
 glass. 
 
 In the assay of brass and German silver, the quantity 
 employed is fluxed with its own weight of borax-glass, 
 but only requires ten times its weight of assay-lead. The 
 operation is commenced as before, but the globule is kept 
 somewhat longer in rotation (always keeping the flame 
 directed only on to the borax glass), so as to allow the zinc 
 present to be completely volatilised, which is evident when 
 the surface of the silver-lead becomes bright, on which the 
 heat is increased for a few moments to expel the last traces 
 of that metal, and the silver-lead thus obtained is concen- 
 trated and cupelled as before. 
 
 The silver globule obtained from the cupellation of 
 substances rich in copper generally requires the addition 
 of a small quantity of lead and re-cupellation (as before 
 described), in order to insure its freedom from copper. 
 
 b. Containing Tin: argentiferous tin, bronze, bell and 
 gun metal, bronze coinage, fyc. Alloys of silver with other 
 metals containing tin do not admit of being cupelled, since 
 the oxide of tin formed by the oxidation of that metal is 
 not absorbed by the bone-ash of the cupel along with the 
 litharge ; it consequently remains upon the surface of the 
 cupel, and if present in any quantity interferes with the 
 operation. As tin is not volatile when heated on charcoal 
 either in the oxidising or reducing blowpipe flame, it 
 cannot be so dissipated, and, in consequence, the entire 
 amount of tin contained in any alloy under examination 
 must be removed by oxidation or scorification from the 
 silver-lead, previous to its being submitted to cupellation. 
 
 For this purpose, 1 part of the stanniferous alloy is 
 fluxed with from 5 to 15 parts of granulated assay lead (ac- 
 cording to the amount of copper suspected to be present in 
 the alloy), O5 part anhydrous sodium carbonate, and 0'5 
 
FORBES'S BLOWPIPE ASSAY. 735 
 
 part pulverised borax-glass, made up as usual in a soda- 
 paper cornet, and the whole at first gently heated in the 
 reducing flame until the soda-paper is charred and the 
 alloy has afterwards united with the lead to form a single 
 globule, whilst the borax and soda have combined as a 
 glass or slag in which the soda prevents the easily oxidis- 
 able tin becoming oxidised to any extent before a perfect 
 alloy has been formed with the lead, which then contains 
 the whole of the silver. 
 
 As soon as this is effected, the blowpipe flame is altered 
 to an oxidising one, and the metallic globule is kept at 
 the point of the blue flame, which should touch it so as to 
 cause the tin to become oxidised and be at once taken up 
 by the glass surrounding it. 
 
 Should, however, it be seen that minute globules of 
 metallic tin made their appearance on the outer edge of 
 the slag or glass,* the operation must be at once discon- 
 tinued, and the assay allowed to cool ; after cooling the 
 metallic globule is detached from the slag surrounding it, 
 and, being placed in a cavity on charcoal, is fused in the 
 reducing flame along with a small piece of borax-glass 
 and afterwards treated with the oxidising flame exactly as 
 before (and if necessary, which is seldom the case unless 
 when treating argentiferous block-tin, this operation may 
 again require to be repeated), until it is seen that the 
 surface of the metallic silver-lead globule does not any 
 longer become covered with a crust or scales of tin oxide, 
 but presents a pure and bright metallic surface. 
 
 The silver-lead globule is now quite free from tin, and 
 can be cupelled and the amount of silver estimated as 
 usual. 
 
 c. Metallic alloys containing much antimony, tellurium, 
 or zinc : antimonial silver and argentiferous antimony, tel- 
 luric silver, and argentiferous zinc. Alloys of antimony 
 with silver when treated on charcoal in the oxidising 
 flame give off all their antimony, leaving the silver behind 
 as a metallic globule having a frosted external appearance ; 
 
 * This occurs when the flux has become so saturated with tin oxide that it 
 cannot take up any more. 
 
736 THE ASSAY OF SILVEft. 
 
 telluric silver, on the contrary, however, when treated 
 in a similar manner, only evolves a part of its tellurium, 
 and even after cupellation with lead a small amount 
 of tellurium generally remains behind alloyed with the 
 silver. 
 
 All these compounds may be assayed as follows : 
 
 One part of the alloy is placed in a soda-paper cornet 
 along with 5 parts of granulated assay lead, and 0-5 part 
 of pulverised borax-glass, and fused in the reducing flame 
 until the globule and slag are well developed ; the oxidising 
 flame is now directed on to the globule, causing the 
 whole of the zinc, along with most of the antimony and 
 part of the tellurium, to volatilise before the lead com- 
 mences oxidising. The last traces of antimony are re- 
 moved with some difficulty, during which operation some 
 portion of the lead becomes oxidised. On cooling, the 
 globule is separated from the slag and concentrated upon 
 a coarse bone-ash cupel as usual, and if 110 tellurium were 
 present in the concentrated silver-lead, this may now be 
 cupelled as usual. 
 
 If tellurium be present as is seen by the concentrated 
 globule of silver-lead possessing a dark-coloured exterior 
 it must be re-melted with 5 parts of assay lead and again 
 concentrated ; and these operations, if necessary, must be 
 repeated until the surface of the concentrated globule is 
 found to be clean and bright, as is usual with pure silver- 
 lead, when it may be cupelled fine and the silver globule 
 weighed. 
 
 It sometimes happens, even after all these precautions 
 have been taken, that the silver globule after cupellation 
 shows a crystalline, greyish-white, frosted appearance, from 
 its still containing tellurium ; in such cases its own weight 
 of assay lead (in one piece) should be placed beside it on 
 the cupel, melted together, and the globule again cupelled 
 fine on another part of the surface of the same cupel. In 
 assaying substances very rich in tellurium the results ob- 
 tained are, however, not very satisfactory, and may be as 
 much as one or two per cent, too low, even after employ- 
 ing all precautions. 
 
FOKBES'S BLOWPIPE ASSAY. 737 
 
 d. Compounds of Silver with Mercury : arquerite, native 
 and artificial amalgams and argentiferous mercury. The 
 assay of these compounds is very simple. A weighed 
 quantity of the liquid or solid amalgam is placed in a 
 small bulb-tube, and heated over the lamp very gradu- 
 ally in order to avoid spirting and to allow the mercury 
 to volatilise quietly ; * the heat is increased by degrees 
 as long as any mercury is driven off, and the residue is 
 heated for some time at a red heat in order to drive off 
 as much mercury as possible without fusing the glass or 
 causing the residual silver to adhere to it. The mercury 
 expelled condenses above the bulb on to the upper 
 part of the tube, and by gently tapping will collect in 
 globules, which, by carefully turning the tube, unite and 
 can be poured out of the tube ; after which the silver, 
 left behind as a porous mass, may be removed from the 
 tube, and after being fluxed with an equal weight of 
 granulated assay lead and half its weight of borax-glass, 
 must be fused on charcoal in the reducing flame, and the 
 button, on cooling, cupelled as usual. Should, however, 
 much copper have been present in the amalgam, a pro- 
 portionately larger amount of assay lead is required to be 
 added. 
 
 When the argentiferous residue is extremely small, as 
 is often the case when assaying argentiferous mercury, 
 this may adhere firmly to the glass of the tube. On such 
 occasions this part of the tube must be cut off with the 
 adherent residue, and the whole fused in a strong reducing 
 flame along with its own weight of granulated assay lead, 
 and with half its weight of anhydrous sodium carbonate. 
 Upon cooling, the globule of silver-lead thus obtained is 
 cupelled as usual. 
 
 e. Compounds chiefly consisting of Iron: argentiferous 
 steel ; cast iron ; bears from smelting furnaces. Compounds 
 consisting principally of iron with a small percentage of 
 silver, although occasionally produced in the arts inten- 
 
 In the case of solid amalgams, which often spirt very violently, this 
 may be obviated by wrapping the assay in a small piece of tissue paper, and 
 heating it in a blowpipe crucible, when all the mercury is given off quietly, 
 leaving the silver behind. 
 
 SB 
 
738 THE ASSAY OF SILVER. 
 
 tionally, as, for example, the so-called silver-steel, are 
 commonly found on the blowing-out of furnaces used in 
 the smelting of silver and copper ores, and are frequently 
 rich in silver, as is the case with the bears from the silver 
 furnaces at Kongsberg in Norway. An alloy of iron with 
 silver is occasionally also found appearing in small quanti- 
 ties on the surface of melted silver in the process of cast- 
 ing, and in some cases at least this may be due to the action 
 of the melted silver on the iron rods used for stirring up 
 the molten metal. 
 
 As iron cannot be made to alloy itself with lead before 
 the blowpipe, it becomes necessary to extract the silver by 
 a more indirect process than is used in the case of other 
 alloys containing that metal. In order to remove the iron 
 the alloy must first be converted into iron and silver sul- 
 phide, and to effect this the iron or steel must be reduced 
 to powder, or fragments none greater than about a quarter 
 of a grain in weight ; for which purpose steel when har- 
 dened may require to be softened previously. 
 
 One part of the finely divided iron or steel is now mixed 
 with O75 part sulphur, eight parts granulated assay lead, 
 and one part pulverised borax-glass : the mixture after 
 being placed in a soda-paper cornet is carefully fused in a 
 cavity on charcoal in the reducing flame, until the whole 
 appears as a fluid globule containing both the lead and 
 iron in combination with the sulphur. Without removing 
 either this globule or the glass surrounding it from the 
 charcoal, an amount of borax-glass in one or more frag- 
 ments (in all about equal in weight to the original amount 
 of iron employed), is now added (in order to combine with 
 and slag off the whole of the iron), and fused along with the 
 former globule, after which the whole is submitted to a 
 strong oxidising flame until the impure lead globule shows 
 itself protruding from the slag. 
 
 The charcoal is then inclined, so that the lead is alone 
 subjected to the action of the outer flame, in order to vola- 
 tilise the sulphur, and at the same time oxidise the iron 
 which goes into the slag ; this operation is continued until 
 the globule of lead appears with a bright metallic surface ; 
 
ASSAY OF ALLOYS OF SILVER AND COPPER. 739 
 
 should it on cooling, however, be found to possess a black 
 colour, and to be brittle, it must be still further oxidised 
 as before described. 
 
 The silver-lead thus obtained will now be found to 
 contain all the silver, and at the same time to be free from 
 both iron and sulphur, and can be cupelled as usual. 
 
 No notice is here taken of alloys of silver and gold, 
 since these metals cannot be separated before the blowpipe 
 by any process yet known ; and in all cases where gold 
 may be present in an alloy treated as here directed for 
 obtaining its contents in silver, the gold also will be found 
 to follow along with the silver, and must be parted from 
 that metal by the wet method, in order to enable the 
 true amount of silver present in the substance to be ascer- 
 tained. 
 
 Alloys of Silver and Copper. A quantity of from 1 
 to 2 grms. is placed in a small flask or beaker, treated 
 with a sufficient quantity of pure nitric acid of a moderate 
 strength, covered with a watch-glass and digested at a 
 gentle heat till totally dissolved. The solution is then 
 diluted with water, and the silver precipitated with hydro- 
 chloric acid, which is added drop by drop as long as any- 
 thing is thrown down. It is then allowed to digest at a 
 very gentle heat till the supernatant liquid is clear. The 
 liquid is then poured upon a dried and counterpoised filter ; 
 the silver chloride is stirred up Avith a little water, and 
 brought upon the filter, the glass being perfectly rinsed out 
 by aid of the washing-bottle. A few drops of nitric acid 
 may be usefully added to the rinsing-water. The washing 
 is afterwards continued with pure water till the droppings 
 have no longer an acid reaction. The filter with its con- 
 tents is dried, and from its weight that of the silver is 
 calculated. 
 
740 
 
 CHAPTEE XVII. 
 
 THE ASSAY OF GOLD. 
 
 FOR the purposes of assay, all substances containing gold 
 may be divided into two classes, as in the case of silver. 
 
 The First Class comprises ores containing gold in a 
 mineralised form. 
 
 The Second Class, comprises all alloys of gold, native or 
 artificial. 
 
 CLASS I. Minerals. 
 
 Graphic Tellurium, (AuAg) Te 3 , containing 30 per cent. 
 of gold, and 10 per cent, of silver. 
 
 Foliated Tellurium, (Pb,Au,Ag) 2 (Te,Sb,S) 3 containing 
 about 9 per cent, of gold. 
 
 CLASS n. Alloys of Gold. 
 
 Native Gold, Au Ag, containing 65-99 per cent, of 
 gold. 
 
 Palladium Gold, AuPd, containing about 86 per cent, 
 of gold. 
 
 Rhodium Gold, AuEh, containing 59-66 per cent, of 
 gold, 
 
 Gold Amalgam, (Au,Ag) 2 HgJ, containing 38 per cent, 
 of gold. 
 
 Artificial Alloys, Gold coin, jewellery, &c. 
 
 Of the foregoing list, native gold alone occurs in nature 
 in sufficient abundance to acquire any great commercial 
 value. It is commonly found in a quartzose gangue, and 
 nearly always associated with one, or more, of the following 
 
. NATIVE GOLD. 741 
 
 minerals : iron and copper pyrites, mispickel or arsenical 
 pyrites, blende, galena, many antimonial minerals, and 
 nearly all the primitive rocks. All auriferous slags, amal- 
 gamation residues and tailings, belong to this class. 
 
 If silver or platinum is associated with the gold in the 
 mineral subjected to assay, it will be found combined with 
 the gold obtained by cupellation ; and all gold so obtained 
 must be submitted to the ' parting process,' which see 
 further on. 
 
 When gold is associated in quantity with quartz, its 
 percentage can be approximately ascertained in the same 
 manner as that of pure tin-stone when mixed with quartz 
 (see page 545). If possible, a fragment of the gold must 
 be detached from the quartz, and its specific gravity 
 taken : if this be not possible, and the gold is nearly 
 fine, the number 19 may be adopted. It is better, how- 
 ever, to estimate experimentally the specific gravity of 
 both quartz and gold. 
 
 Native Gold and Aurides of Silver (Native) are found 
 in variously contorted and branched filaments, in scales, 
 in plates, in small irregular masses, in the crevices or on 
 the surface of common ferruginous and other quartz. 
 
 The author has received two specimens of gold one 
 from Wales, and the other from the Britannia Mine, Devon 
 and found both to be absolutely fine gold. 
 
 Artificial Alloys of Gold. The only one of these alloys 
 which will be specially noticed here is the standard gold 
 of this realm. It is composed of 22 parts of fine gold and 
 2 parts of alloy, constituting 22-carat or standard gold. 
 
 ASSAY OF GOLD ORES.* 
 
 The assay of gold ores embraces the following steps : 
 1. Preparation of the sample. 2. Collection of the 
 gold in a button of metallic lead. 3. Cupellation of the 
 lead button, by which the lead is oxidised and absorbed by 
 
 * For some portions of the following method of assaying gold ores, the 
 Editor is indebted to Mr. T. M. Blossom, who has had great experience in the 
 laboratory of the School of Mines, Columbia College, U.S.A. (See ' Chemical 
 News,' Nos. 607, 609, 628, 635, 636.) 
 
742 THE ASSAY OF GOLD. 
 
 the cupel, leaving behind a bead of the gold and silver. 
 4. Weighing the bead. 5. Inquartation, parting, and 
 cupellation of the gold residue. 6. Weighing the gold 
 bead. 
 
 1. Preparation of the Sample. It is essential, in the 
 first place, to obtain a fair average sample of the ore, 
 otherwise the results of the assay may be commercially 
 worthless. Selection must be left to the judgment of the 
 assayer. The sample must be dried, if necessary, care 
 being taken not to roast it. It must then be pounded in 
 an iron mortar and passed through a sieve of eighty meshes 
 to, the linear inch. If any native metal, in the form of 
 scales or filaments, remain upon the sieve, take the weights, 
 separately, of what has passed through and of what is left 
 upon the sieve. The latter must be assayed according to 
 * Assay of Alloys,' and the result referred to the whole 
 amount of ore. It is essential that the whole of the sample, 
 except the malleable portion, be passed through the sieve. 
 Mix thoroughly the sifted ore. 
 
 2. The Collection of the Gold and Silver in a button of" 
 metallic lead is effected in a crucible, or in a scorifier, 
 whence two methods of assay : 
 
 ~\a) Crucible assay. (b) Scorification assay. 
 
 The former is applicable to all ores ; the latter is 
 lima ted /practically, by the small size of scorifiers, to the 
 richer ores. 
 
 -(a) .Crucible Assay. An ore of gold and silver is com- 
 posed of precious metal, gangue, and oxides, sulphides, 
 &c., of foreign metals. To collect the precious metals in 
 a button of lead, the ore is mixed with litharge, suitable 
 fluxes, an oxidising or a reducing agent, and fused in the 
 crucible. Litharge is reduced to metallic lead ; the latter 
 seizes upon the precious metals and collects in a button at 
 the bottom of the crucible, while the foreign materials 
 form, with the fluxes, a fusible slag above the lead 
 button. The crucible is broken when cold, and the malle- 
 able button detached from the slag by hammering on an 
 anvil. 
 
GOLD OKES. 743 
 
 The Charge. The weight of ore taken for an assay 
 depends upon its supposed richness or poverty, since it is 
 required to obtain finally a bead of precious metal of con- 
 venient size for weighing, and, at the same time, neither 
 too large for cupellation (vide cupellation) nor so small 
 as seriously to affect the calculated results, through losses 
 sustained 'in the assay. As a rule, it is usual to take 
 500, 1000, or 2,000 grains for gold ores. 
 
 The ores require the following reagents : Litharge, 
 sodium carbonate, and one of the reducing agents, argol 
 and charcoal, or an oxidising agent, as nitre, with invariably . 
 a cover of salt one-quarter to half an inch in depth. 
 Borax, silica, and other reagents are very useful at times, 
 but no general rule can be given for their employment. 
 This matter must be left to the judgment of the assay er, 
 guided by the known properties of the reagents and by 
 the composition of the ore. It is well, in this connection, 
 to bear in mind the principle that for basic impurities 
 an acid flux is needed, and for an acid gangue a basic 
 flux. As a rule, employ a weight of litharge twice that 
 of the ore, and of sodium carbonate, the same as of ore. 
 These proportions also may be modified to advantage, 
 according to the composition of the ore. The propor- 
 tion of nitre, or of reducing agent, depends upon the re- 
 ducing power of the ore, hence it is variable in every case. 
 These reagents are added to control the size of the lead 
 button. 
 
 Size of the Lead Button. There are two limits to the 
 size of the button : (1) it must be large enough, or, in 
 other words, enough litharge must be reduced throughout 
 the mass to collect all the precious metal, and, at the same 
 time, (2) there should not be a useless excess of lead, 
 which would occasion loss of silver in the subsequent 
 cupellation. 
 
 These two conditions cannot always be fulfilled, but 
 in this case the latter must be sacrificed. It has been 
 found that a button of 225 grains is the best size 
 for a weight of ore from 500 to 2,000 grains. This is 
 also a convenient size for a cupellation. A button 
 
744 THE ASSAY OF GOLD. 
 
 that is too large for cupellation can, however, always be 
 reduced in size by scorifying. 
 
 These requirements necessitate, in many instances, a 
 preliminary assay of the ore to estimate its reducing 
 power. The reducing power of an ore is due to the 
 presence of sulphur, arsenic, antimony, zinc, &c., but 
 generally to sulphur contained in pyrites, &c. 
 
 Preliminary Assay of Ore. 
 
 Charge. Ore . . . . .20 grains. 
 
 Litharge . . ; .' ' .250 
 
 Sodium Carbonate . .100 
 
 Salt . . . . . Cover. 
 
 A duplicate assay need not be made. 
 
 Warm the crucible before placing it in the fire, which 
 should be perfectly bright, and should be urged to effect 
 complete fusion in the shortest possible time. When the 
 contents of the crucible are in quiet fusion, it must be 
 withdrawn, to prevent further reducing action of the 
 furnace gases. Tap the crucible gently, and when cold 
 break it open. 
 
 Three cases may arise here. Twenty grains of ore may 
 yield 
 
 1. No lead, or less than thirty grains. 
 
 2. Thirty grains lead. 
 
 3. More than thirty grains. 
 
 The reducing power is stated thus : 20 grains ore 
 =# grains lead. Let us suppose that we shall take for the 
 regular assay 150 grains of ore, and that the reducing 
 power is found to be 20 grains ore=15 grains lead: 
 150 grains (about) ore will in this case reduce 112-5 
 grains lead, and as the required button is 225 grains, we 
 must add enough argol, or charcoal, to reduce 112-5 grains 
 in addition ; taking argol as 5*6, we shall require 112'5-^5'6 
 grains, or 20 grains, or charcoal 112-5-*-28 = 4 grains. 
 
 If the reducing power correspond to the third case, a 
 similar calculation will indicate how much nitre is needed 
 
GOLD ORES. 745 
 
 to oxidise part of the sulphur, arsenic, or other reducing 
 agent, and thus prevent the reduction of a button larger 
 than 225 grains. One part of lead requires 0*25 part of 
 nitre to oxidise it. In the second case, 150 grains of 
 ore would reduce a button of 225 grains, and neither argol 
 nor nitre would be required. 
 
 The character of slag obtained in the preliminary assay 
 may also suggest some modification of the regular charge. 
 If it be earthy, for instance, we would add borax-glass or 
 silica. Experience will often enable the assayer to judge 
 of the reducing power with sufficient exactness, without an 
 extra assay. Cases will arise, however, in which he must 
 make a preliminary assay, or roast the ore, and it is 
 always best to roast when there is a large amount of 
 sulphur in the ore. He will then have an ore of no 
 reducing power. (Case I.) 
 
 Ores to be Roasted. Ores containing a large amount of 
 sulphur, or arsenic, antimony, or zinc, should always be 
 roasted. In the former case, if the ore be not roasted, 
 there will be danger of the formation of oxysulphides, 
 which are very fusible, but are not decomposed even at a 
 white heat, and enter the slag carrying silver with them. 
 A large quantity of nitre also subjects the contents of the 
 crucible to the liability of boiling over : even should this 
 mishap not occur, the great evolution of nitrous and sul- 
 phurous vapours puffs up the mass throughout the cru- 
 cible, and the globules of lead afterwards reduced may be 
 left adhering to the sides, not being washed down by the 
 retreating charge. Arsenic and antimony produce arseni- 
 ates and antimoniates, which carry silver into the slag. 
 Zinc increases the loss of silver by volatilisation and also 
 in the slag. 
 
 Roasting the Ore. The ore may be roasted conveniently 
 in a cast-iron pan over the crucible furnace. There ought 
 to be a hood over the furnace, to carry off the fumes of 
 sulphurous, arsenious, and other acids. The pan should 
 be covered with a coating of red ochre, or, still better, 
 of chalk. The former is put on wet with a brush. An 
 excellent even coating of the latter may v be obtained by 
 
746 THE ASSAY OF GOLD. 
 
 making a chalk paste, of proper consistence, in the pan, 
 and turning with the hand so as to spread the paste while 
 the pan is being held over the fire to dry. The coating 
 prevents a loss of ore and injury to the pan through the 
 sulphides attacking the iron. 
 
 The weighed sample must be spread over the pan and 
 stirred with a bent wire until all danger of fusion is past. 
 The pan must be heated gradually at first, not above a 
 dull red heat for some time, and may be brought, finally, 
 to a full red, or higher heat. Too high a heat at the 
 outset might cause the fusion of sulphides and the forma- 
 tion of matter troublesome to roast. A rapid disengage- 
 ment of arsenic, antimony, or zinc would cause, also, a 
 mechanical loss of silver, by carrying it off in their vapours. 
 Should fusion take place, it is better to weigh out a fresh 
 portion of ore and roast again with more care. If this be 
 not done the fused portions must be pulverised in a mortar 
 and re-roasted. Generally, in roasting sulphides, the 
 operation may be considered finished when, after keeping 
 the pan at a full red heat for some time, no more fumes of 
 sulphurous acid can be perceived. As there is danger, 
 however, of the formation of sulphates, especially if copper 
 pyrites be present, it is best always to mix some ammo- 
 nium carbonate with the ore, after the fumes have ceased, 
 and to cover the pan. The ore is thus confined in an 
 atmosphere of ammonium carbonate, which decomposes 
 the sulphates with formation of volatile ammonium sul- 
 phate. 
 
 Arsenic and antimony require the addition of fine 
 charcoal, to reduce any arseniates and antimoniates that 
 may have been formed during the roasting. Care must 
 be taken to burn out all the charcoal. 
 
 If the ore contain a very fusible sulphide, as antimony- 
 glance or galena, it may be mixed with some fine sand 
 previous to roasting 
 
 The roasting of ores may also be done in the muffle in 
 an earthen saucer. 
 
GOLD ORES. 
 
 747 
 
 Fusion. 
 
 The charge prepared according to the foregoing direc- 
 tions is thoroughly mixed and placed in a crucible, which 
 it should not more than two-thirds fill. A hot fire should be 
 employed, and the crucible removed when complete fusion 
 has taken place. This ought to require from twenty to 
 twenty-five minutes. The crucible is tapped, as usual, and 
 broken when cold. 
 
 (b) Scarification Assay. The reagents necessary for a 
 scorification assay are test-lead and borax-glass. The ore 
 is mixed with these in suitable proportions, the mixture 
 put in a scorifier and fused in a muffle. The operation 
 affords an alloy of lead with the precious metals, and a 
 slag composed of litharge with the impurities and gangue 
 of the ore. The assay might be made with lead only, 
 but it is advantageous to add some borax. There will be 
 required of lead only enough to render the slag liquid and 
 to furnish lead for the button. The proportions of both 
 lead and borax will vary, and should be greater in pro- 
 portion as the gangue and metallic oxides are more diffi- 
 cult of fusion. The following table exhibits the propor- 
 tions found by experience to be best adapted to the 
 different gangues. The proportions are referred to one 
 part of ore. 
 
 Parts 
 Test-Lead 
 
 Parts 
 Borax 
 
 8 
 8 
 5-6 
 16 
 16 
 12-16 
 10-15 
 10-15 
 
 0-25-1-00 
 0-15 
 0-10-0-50 
 0-10-1-00 
 0-10-0-15 
 0-10-0-20 
 0-10-0-20 
 
 Character of Gangue 
 
 Quartz ose 
 
 Basic (Fe 2 O 3 ,AL0 3 CaO), &c 
 
 Galena . 
 
 Arsenical 
 
 Antimonial 
 
 Fahlerz 
 
 Iron pyrites 
 
 Blende . 
 
 No preliminary roa.sting is required. As in the cru- 
 cible assay, the weight of ore taken depends very much 
 upon its richness, but is generally a third, sixth, or tenth 
 of an assay ton.* If one scorifier will not contain the 
 charge, it is best to weigh equal fractional parts for the 
 
 * An assay ton is about 450 grains. 
 
748 THE ASSAY OF GOLD. 
 
 number required, rather than to weigh the whole charge 
 and roughly divide it between the scorifiers. In case of 
 an accident to one of the scorifiers, the known loss can 
 easily be restored, if the former course be followed. The 
 exact parts that lead and borax perform in scorification 
 can be understood best from a description of the operation 
 in detail. 
 
 Three distinct periods maybe noted in the working : 
 1. Eoasting ; 2. Fusion ; 3. Scorification. A strong- 
 heat is maintained, at first, in order to melt the lead. 
 'This is effected by closing the muffle and regulating the 
 draught. As soon as the lead is fused the muffle is 
 opened, and the ore is seen floating upon the surface of 
 the lead. 
 
 1. The roasting now commences, and is continued at 
 a moderate heat until no more fumes are seen, and the 
 ore has disappeared. 
 
 2. The heat should now be raised in order to fuse 
 thoroughly all the material in the scorifier. When the 
 fusion is complete, clear white fumes of lead may be seen 
 arising from the scorifier, there is an intermittent play of 
 colours across the bright surface of the lead, and the slag 
 produced encircles the metallic bath like a ring. The 
 borax plays an important part just here, by giving liquidity 
 to the slag, thus permitting it to be thrown to the side as 
 fast as formed, exposing a clear surface of lead -for oxida- 
 tion. If borax be not added and the ore contain a diffi- 
 cultly fusible gangue, the scoriae will float in detached 
 masses over the lead, impeding the oxidation, until suffi- 
 cient litharge has been formed to give it liquidity. 
 
 3. When the fusion is complete, the heat may be 
 lowered and the third period of scorification continued 
 until the ring of slag, ' which is continually growing 
 smaller, closes over the residue of metallic lead. The heat 
 .should again be raised, to liquefy the slag and allow the 
 metallic lead to settle, after which the scorifier is removed 
 from the furnace. If the button is wanted immediately 
 for cupellation, the contents of the scorifier may be poured 
 into an iron or copper mould coated with reel ochre. It 
 
CUPELLATIOiV. 749 
 
 is thus soon cooled. Otherwise allow the contents to 
 solidify in the scorifier, and break this for the button 
 when cold. Hammer the button as usual. 
 
 The whole assay occupies about thirty-five minutes. 
 
 In making the charge, it is customary to mix the ore 
 with a part of the test-lead and with borax, and to cover 
 this in the scorifier with the rest of the lead. This will 
 prevent loss of ore prior to fusion. Too much borax 
 should not be added at first. If a large amount is needed, 
 it is better to mix only a portion with the ore ten grains, 
 for instance and to introduce the rest wrapped up in 
 paper, as needed during the operation. Some of it may 
 be reserved for the final heating after the lead is slagged 
 over. If too much were added at first, the lead would be 
 slagged over before the necessary reactions had taken place. 
 
 The Lead Button. The lead button submitted to cupel- 
 lation must be malleable and of the proper size for the 
 cupel. A good cupel will absorb its own weight of 
 litharge, but it is better to use a cupel one-quarter to a 
 third as heavy again as the lead-bottom. The cupels in 
 ordinary use weigh about 250 grains, hence a button of 
 180 to 230 grains is the proper size for them. If the 
 button be too large it may be reduced in size by scorifica- 
 tion. In case of doubt it is better to scorify, since there is 
 less loss in this operation than in cupellation. A brittle 
 button may be due to the presence of sulphur, arsenic, 
 antimony, zinc, or litharge. In either case it must be 
 scorified before cupellation, and with test-lead if necessary. 
 If the button contain copper it must be scorified until no 
 more copper can be seen on hammering out the button, 
 If nickel be present the button cannot be cupelled. This, 
 however, will occur but rarely. 
 
 3. Cupellation. The lead button obtained by the fore- 
 going operations is next cupelled. This operation is 
 similar in many respects to scorification, but differs from 
 it in that the scoriae formed are absorbed entirely by the 
 cupel, leaving a pure bead of the precious metals. This 
 property of the absorption of scorise is an indispensable 
 condition in cupellation, so that, unlike scorification, it is 
 
750 THE ASSAY OF GOLD. 
 
 limited to a few substances capable of being absorbed by 
 the cupel. The oxides of lead and bismuth alone can be 
 absorbed in a state of purity, but they can carry along 
 with them certain proportions of other substances which, 
 by themselves, would form infusible scoria. It is thus 
 that we are enabled to get rid of small amounts of copper, 
 iron, arsenic, &c., remaining in the lead button, Bismuth is 
 seldom, if ever, used for this purpose. The proportion of 
 lead required to carry off the different impurities varies 
 according to circumstances. In the case of lead buttons 
 from an assay, there is always an excess of lead, if they 
 have been properly purified. Other cases will be treated 
 under the head of Assay of Alloys, where the necessary 
 tables will be given. 
 
 The operation of cupelling a lead button is conducted 
 t as follows : 
 
 The cupels must be carefully dried before use, and 
 must be free from cracks, which would cause a loss of 
 precious metal. The bottom of the muffle should be 
 covered with sand to prevent injury to it in case of the up- 
 setting of a cupel. A cupel having been selected, it should 
 be wiped carefully with the finger, and all extraneous 
 matter blown out. It may then be placed in the muffle. 
 Before the introduction of the lead button, the muffle 
 should have attained a reddish-white heat, and the cupel 
 should be of the same temperature. This being attained, 
 the button is placed in the cupel with a pair of forceps, 
 gently, so as not to injure it. The muffle should now be 
 closed, either by a door or by a piece of lighted charcoal, 
 to bring the fused button to the same temperature. This 
 done, the muffle is opened and air allowed to enter ; the 
 button, which at first appears bright and uncovered, is 
 soon covered with a film of oxide moving in luminous 
 patches over the surface, and being continually thrown 
 toward the edge, where it is absorbed by the cupel. White 
 fumes of lead arise, and the button is surrounded by a 
 ring of the absorbed litharge, which continually widens 
 until it reaches the edge of the cupel. The button thus 
 gradually diminishes in size by oxidation and absorption, 
 
CUPELLATIOtf. 751 
 
 and becomes more convex ; the luminous patches become 
 larger and move more quickly ; and when, finally, the last 
 of the lead is absorbed, the button appears to revolve 
 rapidly on its axis, becomes very brilliant, and is suffused 
 with all the tints of the rainbow ; the movement is sud- 
 denly checked, the button becomes dull for a few instants, 
 and then presents the appearance of the pure precious 
 metals. The latter part of the operation is called the 
 4 brightening ' of the button. Should the bead be large, 
 and composed, in great part at least, of silver, it must be 
 removed slowly and gradually from the furnace, to pre- 
 vent loss by ' spitting,' which might happen if the cupel 
 were removed at once from the muffle. When a cupel is 
 withdrawn directly after brightening, the button is liable 
 to be covered by mammillated and crystalline protuber- 
 ances, and is then said to have ' vegetated ' or ' spit.' 
 Portions of the metal are sometimes thrown off and lost. 
 Whatever the cause may be, this does not occur if the 
 button be withdrawn gradually, so as to permit a slow and 
 gradual cooling. In case the bead be very large, say 
 2 to 5 grains, it is well to cover it with a hot cupel, 
 which will retard the cooling and prevent the loss of small 
 particles that may be thrown off despite all care. If the 
 bead is not larger than the head of an ordinary pin, the 
 danger of vegetation is slight, and no great precautions 
 need be taken in its removal. 
 
 It is well to raise the heat of the furnace just before 
 the brightening, or to push the cupel into a hotter part 
 of the muffle, in order to aid the brightening and get rid 
 of the last traces of lead, which are somewhat difficult to 
 remove. The button should also, for the latter purpose, 
 be heated strongly a few seconds after brightening. 
 
 Silver is sensibly volatile at a high heat ; the loss of 
 gold is slight, and may be disregarded. In the assay of 
 ores this loss need not be considered, but in bullion assay 
 a correction is necessary, for which a table will be given 
 under Assays of Alloys. The loss of silver increases with 
 the temperature, but we must avoid, in cupellation, the 
 two extremes of a high heat and quick work and a low 
 
752 CUPELLATION. 
 
 heat with prolonged work. Of the two extremes the latter 
 is worse. The following are indices of favourable work- 
 ing : The muffle is reddish-white, the cupel is red, the 
 fused metal very luminous and clear, the lead fumes rise 
 slowly to the top of the muffle, and the litharge is com- 
 pletely absorbed by the cupel. 
 
 The heat is too great when the cupels are whitish, when 
 the fused metal is seen with difficulty, and the scarcely 
 visible fumes rise rapidly in the muffle. 
 
 The heat is too low when the fumes are thick and fall 
 in the muffle, and when the un absorbed litharge is seen 
 forming lumps and scales about the button. 
 
 The degree of heat should bear some relation to the 
 richness of the alloy, and may be greater according as 
 the lead is poorer in silver. By bearing this in mind the 
 assayer may often hasten the operation without detriment 
 to the assay. The draught of air through the muffle must 
 also be regulated. Too strong a current cools the cupel 
 and oxidises the lead faster than it can be absorbed, thus 
 endangering the assay. Too slow a current prolongs the 
 operation and increases the loss by volatilisation, In 
 ordinary work this matter, however, occasions very little 
 trouble. 
 
 It happens sometimes that the material in the cupel 
 becomes solidified in the midst of an operation, stopping 
 all further action. This disaster is called the ' freezing ' of 
 the button, and is occasioned by the following conditions : 
 a production of litharge more rapidly than it can be ab- 
 sorbed by the cupel, and infusible scoriae, due to a cold 
 furnace, or to an excess of foreign oxides. In either case 
 the scoriae gradually extend over the surface of the button 
 until it is entirely covered, when further movement ceases. 
 This disaster may sometimes be prevented or remedied by 
 raising the heat of the muffle ; if this fails, or if the acci- 
 dent be due to foreign oxides, an addition of pure lead 
 must be made to the assay ; in either case, the results are 
 unreliable. 
 
 An assay that has passed well, furnishes a bead well- 
 rounded and clear, crystalline below, and readily to be 
 
WEIGHT OP MINUTE SPHERES OF GOLD. 753 
 
 detached from the cupel. If the bead contain lead, it is 
 brilliant below, and does not adhere at all to the cupel. 
 If the bead exhibit rootlets passing down into the sub- 
 stance of the cupel, the results are inaccurate. 
 
 ESTIMATING THE WEIGHT OF MINUTE SPHERES OF GOLD. 
 
 M. G. A. Gozdorf has given, in the ' Chemical News ' 
 for 1886, an ingenious plan for the production and measure- 
 ment of gold and other minute metallic spheres to esti- 
 mate their weight. 
 
 In making assays for gold where the amount of gold 
 is very small a little silver is required in which the gold 
 may be collected. As nearly all commercial litharge con- 
 tains silver it is rarely necessary to add any separately for 
 this purpose. 
 
 Having obtained a prill in which the amount of gold is 
 a third or less than the silver, the prill is boiled in dilute 
 nitric acid in a porcelain capsule to dissolve the silver, and 
 where the amount of gold is more than 1 dwt. to the ton 
 a second boiling in strong nitric acid should be given. If 
 care is taken in using dilute acid at first and boiling gently 
 the gold will be left in one piece of a nearly black colour. 
 The acid is now decanted off and the gold washed two or 
 three times in distilled water. The gold may be now placed 
 on an aluminium or other polished metal plate by invert- 
 ing the capsule and leading the last drop of water and the 
 gold with a glass rod on to the plate, the water is drawn 
 off by a piece of filter-paper, and the plate gently heated 
 till dry. 
 
 Having thus obtained the gold in a pure state a bead is 
 made of boracic acid on a platinum wire loop and pressed 
 on the gold while still red-hot ; the gold adheres without 
 difficulty, and by heating the bead before the blowpipe the 
 gold is obtained as an almost perfect sphere. 
 
 Should the resulting sphere of gold be very minute it 
 is better to measure it under the microscope while in the 
 bead, but if large enough to be seen with the naked eye 
 it can be measured more accurately after dissolving the 
 
 3c 
 
754 THE ASSAY OF GOLD. 
 
 boracic acid bead in a watch-glass with hot water and 
 placing the sphere of gold on a glass slide. 
 
 The plan of measuring minute prills of silver and gold 
 to estimate their weight was first introduced by Harkort, 
 who used an ivory scale engraved with two fine lines meet- 
 ing at an acute angle and divided into fifty equal parts 
 (see fig. 141, p. 724). 
 
 According to the fifth edition of Plattner's 'Probir- 
 kunst,' p. 520, Goldschmidt estimates the weight of 
 silver and gold prills by measurement with the micro- 
 scope. 
 
 Harkort and Plattner, in making scales for the esti- 
 mation of the weight of gold and silver prills, weighed 
 prills corresponding to the larger divisions of the scale, 
 and from their weight calculated the weight for the smaller 
 divisions. These prills were taken direct from the cupel, 
 and at the point of contact are flattened ; but as the 
 amount of flattening is not always the same, and hardly 
 varies in extent with the size of the prill, and as the con- 
 verging lines on the scale cannot be very sharply defined, 
 this method is not capable of the same accuracy as where 
 the almost perfect spheres are measured with a microscope. 
 
 No other flux seems to possess advantages equal to 
 those of boracic acid for obtaining a sphere of gold. Borax 
 and other fluxes are so fluid when hot that the gold is very 
 liable to alloy with the platinum wire : this rarely occurs 
 with boracic acid, on account of its great viscosity, even 
 when white-hot. Boracic acid is also easily soluble in 
 water, so that the gold spheres can be separated without 
 loss of time. 
 
 The following rules and figures may be useful to any 
 one wishing to adopt the system here described. 
 
 1. The weight of a sphere increases as the cube of the 
 
 diameter. 
 
 2. The weight of a sphere of any substance of which 
 
 the specific gravity is known is obtained by mul- 
 tiplying the weight of a unit sphere of water by 
 the specific gravity of the substance and the cube 
 of the diameter. 
 
WEIGHT OP MINUTE SPHERES OF GOLD. 755 
 
 Constants for Use with Gramme Weights. 
 
 1. Weight of a sphere of water 0-01 m.m. in diameter 
 
 -0-000,000,000,523,6 of a grm. 
 
 2. Weight of a sphere of gold 0-01 m.m. in diameter= 
 
 0-000,000,010,210,2 of a grm. 
 
 3. Weight of a sphere of gold 0'0# m.m. in diameter, 
 
 a* x 0-000,000,010,210,2 of a grm. 
 
 4. If 20 grms. of ore are taken for assay the number 
 
 of grains of gold per ton is found by x* x 0*008004, 
 in which ^?=the diameter of the sphere of gold in 
 hundredth s of a millimetre. 
 
 Constants for Use with Grain Weights. 
 
 1. Weight of a sphere of water 0*001 inch in diameter, 
 
 0-000,000,132,4 of a grain. 
 
 2. Weight of a sphere of gold 0*001 inch in diameter, 
 
 0-000,002,582 of a grain. 
 
 3. Weight of a sphere of gold 0*00# inch in diameter, 
 
 x* x 0-000,002,582 of a grain. 
 
 4. 200 grains of ore being taken for assay, the num- 
 
 ber of grains of gold per ton is found by 
 x 3 x 0-2045288, in which #==the diameter of the 
 sphere of gold in thousandths of an inch. By 
 taking 978 grains for assay # 3 = grains of gold 
 per ton. 
 
 To test the accuracy of the above figures a compara- 
 tively speaking large sphere of gold from an assay was 
 measured and found to be 0*593 m.m. or -^ of a m.m. 
 in diameter; 59*3 3 x 0*000,000,010,210,2=0-002,129 of a 
 grm. When weighed on a very delicate balance it was 
 found to weigh 0*0021 grm., and as this balance does not 
 indicate beyond the fourth decimal the results may be 
 considered identical. This sphere indicated gold in the 
 sample tried at the rate of 3 oz. 9 dwts. 13 grs. per 
 ton. 
 
 The smallest sphere of gold yet measured was 0*024 m.m. 
 
 3 o 2 
 
756 THE ASSAY OF GOLD. 
 
 in diameter, and by applying the above rule the weight 
 would be 2-4 3 x 0-000,000,010,210,2 x 15-43235 (to convert 
 grammes to grains) = 0- 000, 002,178, or a trifle over two 
 millionths of a grain. 
 
 Spheres of silver may be obtained and measured in a 
 similar manner. The boracic acid acts slightly on the 
 silver, but the quantity dissolved is inappreciable, as the 
 action is not prolonged. The specific gravity of silver 
 being 10-53, the weight of l-100th m.m. would be 
 0-000,000,000,523,6x10-53 = 0-000,000,005,513,508 of a 
 grm. In a test assay made with silver the sphere measured 
 0-57=f$j- of a m.m., from which the weight deduced 
 would be 0-001,020,96 of a grm., the balance showing the 
 weight as 0*0010 of a grm. 
 
 Copper, lead, and other metals cannot be melted in 
 boracic acid on platinum wire without dissolving to a 
 perceptible amount, but may with care be melted in sodic 
 carbonate, and, by dissolving the latter in hot water, the 
 sphere of copper, &c., obtained and measured. 
 
 GENERAL OBSERVATIONS ON THE ASSAY OF GOLD ORES. 
 
 Gold and Copper, Proportion of Lead. The alloys of 
 gold and copper are cupelled like the alloys of gold and 
 silver ; but as copper has a very great affinity for gold, it 
 is necessary to use a larger proportion of lead to insure 
 its oxidation when combined with gold than when united 
 with silver. This proportion varies according to the 
 standard and the temperature. It is admitted that for the 
 same standard there must, under similar circumstances, be 
 twice as much lead used in the cupellation of gold as for 
 that of silver. Thus, 14 parts, at least, ought to be em- 
 ployed in common furnaces for an assay of gold coin 
 which contains 0*1 of copper. There is no inconvenience 
 in employing a little more, as it does not increase the loss 
 of gold. However great the proportion of lead may be 
 that is added to the cupreous gold for the purpose of 
 cupellation, the button retains always a very small quantity 
 of copper, which a fresh cupellation does not free it from, 
 
EXAMINATION ON THE TOUCHSTONE. 767 
 
 .and which occasions what is termed the surcharge. This 
 .surcharge, being very slight, can be neglected in assays of 
 minerals ; but it is necessary to take notice of it in the 
 assay of alloys. But it is known that the presence of 
 silver much facilitates the separation of copper from gold, 
 and it is rare that an alloy of cupreous gold does not con- 
 tain a little silver, which must be separated : and when that 
 is not the case, a small quantity of that metal can be 
 introduced into the alloy, so as to be in about the propor- 
 tion of 3 parts to 1 of gold. When an assay is to be made 
 of an alloy of gold and copper, a sufficient quantity of 
 silver is to be added to fulfil this condition according to 
 the presumed standard, which is estimated approxi- 
 matively by a preliminary assay, and then cupelled with 
 lead. 
 
 Examination on the Touchstone. This method is based 
 upon the fact that the richer an alloy is in gold the more 
 clearly does a streak drawn with it on a black ground 
 present a pure gold-yellow colour, and the less is it 
 attacked by pure nitric acid or by a test acid. This test 
 -acid consists of ninety-eight parts pure nitric acid of 
 1*34 sp. gr. (37 Beaume), two parts pure hydrochloric 
 acid of 1-173 sp. gr. (21 B.), and twenty-five parts 
 distilled water. To judge of the richness of the alloy to 
 be examined, its streak is compared with marks drawn 
 with alloys (the touch-needles) whose richness is accurately 
 known. In order to get correctly the streak of the alloy 
 to be tested, the surface of the metal must first be some- 
 what filed away, since this may be impure, or, as with 
 coins and jewellery, it may have been made somewhat 
 richer by boiling with acid, and the so-called colouring of 
 the goldsmith, and a clean fracture is rarely to be obtained. 
 .Five series of prepared touch-needles are required. The 
 first series consists of copper and gold, and is called the 
 .red series, and the proportion of gold increases by half- 
 carats in the successive needles. The second series, the 
 white series, contains needles of gold and silver, in which 
 the proportion of gold likewise increases by half-carats. 
 The third series, a mixed one, contains needles in which the 
 
758 THE ASSAY OF GOLD. 
 
 quantities of silver and copper are equal, and the propor- 
 tion of gold also increases by half-carats. The fourth con- 
 sists also of needles for a mixed series, in which the silver 
 is to the copper as 2 : 1, and the gold increases by half- 
 carats ; and the fifth is also formed of needles for a mixed 
 series, in which the quantity of silver is to that of the 
 copper as 1 : 2. Moreover, in mints and stamping bureaux, 
 alloys are used which correspond precisely to the legal 
 standards. The testing upon the touchstone begins by 
 determining to which series the alloy to be examined be- 
 longs. Then those touch-needles are rubbed against the 
 stone whose marks most nearly approximate in colour to 
 that of the alloy. The marks must form a thin continuous 
 layer. A drop of pure nitric acid is now placed upon 
 them with a glass rod, and its comparative effect observed. 
 The acid is allowed to work a short time, and then wiped 
 off, in order to see whether the streak appears unchanged, 
 or whether it has more or less disappeared. The test acid 
 above is also used. This is so composed that it does not 
 work at all upon an alloy containing eighteen carats and 
 more of gold, and with such an alloy the streak, after 
 using the acid, will not be wiped off with a fine linen rag, 
 provided that stone and acid had a temperature of 10 to 
 12 C. Pure nitric acid produces almost no effect upon an 
 alloy of fifteen or sixteen carats fine, and over. The test- 
 ing on the touchstone can indeed make no pretension to 
 accuracy, especially where the amount of gold is small, 
 but it yields sufficiently useful results for a preliminary 
 test. It requires, however, a sharp and very practised eye. 
 Moreover, the preparation of the touch-needles is weari- 
 some, as the required proportion is not always quickly 
 reached, nor are good malleable alloys always obtained. 
 The touchstone, therefore, is in general only used where 
 frequent gold assays are to be made of alloys varying 
 in richness, or where (as frequently with gold plate) aiL 
 examination on the touchstone will suffice. 
 
 
CUPELLATION OF GOLD AND COPPER. 
 
 769 
 
 TABLE FOR PROPORTION OF LEAD TO BE EMPLOYED IN THE 
 
 CUPELLATION OF GOLD AND COPPER. 
 
 Gold in alloy 
 
 1000 thousandths 
 
 Lead required 
 
 Eatio of lead in 
 the assay to the 
 copper, <fcc. 
 
 900 
 
 
 100 000 1 
 
 800 
 
 16 
 
 80 000 * 1 
 
 700 
 
 22 
 
 70 qqq . -i 
 
 600 
 
 24 
 
 fift 000 1 
 
 500 
 
 26 
 
 no 00 1 
 
 400 v 
 300 
 200 I 
 
 34 
 
 / 56,666 : 1 
 48,571 : 1 
 
 { 4.0 K()f) . 1 
 
 100 
 50] 
 
 
 (37,377 : 1 
 
 Kandelhardt gives the ratio in the following table : 
 
 Gold in 1000 parts 
 
 1000 fine gold 
 980 920 
 920 875 
 875750 
 750 600 
 ' 600 350 
 350 
 
 Quantity of lead required 
 
 8 times the weight of the alloy 
 12 
 16 
 20 
 24 
 28 
 32 
 
 Gold, Silver, Platinum, and Copper. The presence of 
 platinum in an alloy renders- the separation of the oxi- 
 disable metals, more especially copper, very difficult by 
 cupellation. It appears, indeed, that it would be almost 
 impossible to arrive at it, if the alloy of copper contained 
 nothing but gold and platinum. It is necessary that silver 
 be present at the same time. When this metal is absent, 
 it is requisite to add a quantity of it, which ought to be 
 equivalent to double the weight of the gold and platinum 
 united, and cupel at the strongest heat which can be ob- 
 tained in a good muffle with a suitable proportion of lead. 
 This proportion varies much according to the composition 
 of the alloy, and the temperature at which the operation 
 is carried on. 
 
 Experience has shown that the copper can be more 
 completely separated and less silver lost by cupelling at a 
 high temperature, with the least possible quantity of lead, 
 than by employing more lead, and working at a lower tem- 
 perature. M. Chaudet has made several assays, in order 
 
760 THE ASSAY OF GOLD. 
 
 to estimate the proportion of lead required for the cupel- 
 lation of the three following alloys : 
 
 1. 2. 3. 
 
 Gold .... .0-100 0-020 0-005 
 
 Platinum 0-100 0-200 0-300 
 
 Silver 0-250 0-580 0-595 
 
 Copper 0-550 0-200 0-100 
 
 and has found, for the first, that by employing 20 parts of 
 lead the separation is very nearly complete ; but that at a 
 high temperature there is a loss of silver, and in order to 
 render the assay correct it must be cupelled at the latter 
 temperature, with only 14 of lead ; for the second, 8 of 
 lead, at a high temperature ; and for the third, 30 parts 
 of lead are necessary, at the same high temperature of the 
 muffle ; but it is almost impossible to separate all the 
 copper, and no advantage can be obtained by increasing 
 the quantity of lead. When almost the last traces of the 
 copper are separated, the button must be cupelled afresh, 
 with a small quantity of lead ; but a small quantity of 
 silver is nearly always lost. In all cases, in order that no 
 lead shall remain, it is necessary to leave the assay button 
 some few minutes in the muffle after cupellation is finished. 
 The alloys of gold and silver which contain platinum 
 show, either by cupellation or parting, certain characters 
 which prove the presence of the metal. If the assay be 
 not heated very strongly, it does not pass, and the button 
 becomes flat : this effect becomes very sensible when the 
 platinum is to the gold in the proportion of 2 to 100. 
 Under the same circumstances, the nitric acid solution 
 proceeding from the parting is coloured straw-yellow. 
 At the moment an assay of an alloy containing platinum 
 terminates, the motion is slower, and the coloured bands 
 are less numerous, more obscure, and remain a much 
 longer time than when there is no platinum ; the button 
 does not uncover, and the surface does not become as 
 brilliant as that of an alloy of gold or silver, but it remains 
 dull and tarnished. When the assay is well made, it is to 
 be remarked that the edges of the button are thicker and 
 more rounded than in ordinary assays, and it is of a dull 
 white, approaching a little to yellow; and lastly, its sur- 
 
GOLD AND SILVER PARTING. 761 
 
 face is wholly or in part crystalline. These effects are 
 sensible even when the gold does not contain more than 
 01 of platinum. When the alloy contains more than 10 
 parts of platinum to 90 of gold, the annealed cornet pro- 
 duced in the parting process is of a pale yellow, or tar- 
 nished silver colour. 
 
 Gold alloyed with Silver. The separation of gold from 
 silver is termed parting. Parting is not only used to sepa- 
 rate silver from gold, but for the separation of other metals, 
 such as copper, when cupellation does not separate it 
 entirely. Parting by the wet process is carried on by the 
 means of nitric acid, aqua regia, or sulphuric acid. 
 
 When an alloy of gold and silver has been reduced by 
 a flatting mill to very thin plates, it is sufficient that it 
 contains 2^ of silver to 1 of gold in order that the parting 
 may be effected completely by nitric acid, and it takes place 
 much less easily when the silver in the alloy is in larger 
 proportion ; but when this proportion exceeds 3 parts of 
 silver for 1 of gold, then the latter is obtained in leaves so 
 fine that there is risk incurred of losing some in the sub- 
 sequent manipulation, and even by the act of boiling the 
 acid liquid. 
 
 We must always, therefore, when a very exact assay is 
 required, contrive that the alloy shall contain a little less 
 than 3 parts of silver to 1 of gold, a proportion which 
 long experience has demonstrated to be the best. If the 
 alloy contain less than 2-| of silver to 1 of gold, the silver 
 does not wholly dissolve, because there is a part of it so 
 enveloped in the gold that the strongest acid does not act 
 on it.* 
 
 Inquartation. The operation by which the alloy is 
 brought to this standard is termed quartation, or inquar- 
 tation. It consists in fusing the alloy in a cupel, with 2 
 parts of lead and the quantity of fine silver, or fine gold, 
 necessary to bring it to the desired composition. This 
 quantity is estimated according to the approximative esti- 
 mation of the standard of the alloy, which ought to be 
 
 * Pettenkoffer and others have shown that less than two parts of silver 
 will suffice, and be even advantageous. 
 
762 THE ASSAY OF GOLD. 
 
 made either by means of a preliminary assay, as hereafter 
 described, or by means of the touchstone. If we do not 
 employ the whole of the alloy the assay will not be exact, 
 because the gold and silver are not always found dis- 
 tributed in an uniform manner ; at least every time it is 
 not poured into a cold ingot mould. 
 
 Operation. The cupelled and quartated button is 
 flattened on an anvil and annealed, in order to soften it. 
 It is laminated to give it a certain thickness, and is then 
 annealed afresh, and rolled into a cornet or spiral around 
 the quill of a pen. It is necessary that the alloy should 
 be reduced to a suitable thickness, on the one hand, in 
 order that the silver may be dissolved completely ; and, on 
 the other, that the plate of gold may remain whole after 
 the operation. The following is that which experience has 
 proved best. The quantity of matter operated upon, or 
 taken for the assay, should be about 12 grains, and the 
 alloys resulting from these 12 grains, and the silver em- 
 ployed in the inquartation, should be made into a plate of 
 from 18 to 20 lines * in length and 4 or 5 in breadth. 
 
 The cornet for assay is placed in a glass matrass,, 
 capable of containing about three ounces of water ; pure 
 nitric acid is added at different times, and heat applied. 
 When all the silver is dissolved, it is washed by decanta- 
 tion with water ; the matrass is reversed into a small 
 crucible, the cornet falls out and is dried. In this state 
 the cornet is very fragile, and of a dull red colour ; it is 
 annealed in a muffle, and heated gradually without fusion. 
 It becomes thereby much contracted, and acquires a 
 metallic lustre, and so much solidity that it can be weighed 
 without fear of breaking it. Its weight can be ascertained 
 in the assay balance. 
 
 There are many ways of employing nitric acid. For- 
 merly 2-|- ounces (thirty-five times the weight of the alloy) 
 of nitric acid (1*15 sp. gr.) was poured upon the inquar- 
 tated cornet, and boiled gently for fifteen or twenty 
 minutes, the liquid decanted and replaced by 1-J- of acid 
 
 * A line is the f an inch. , 
 
INQUARTATION. 763 
 
 (sp. gr. 1-24 or T26), twenty-four times the weight of the 
 alloy, boiling for twelve minutes, then decanting and wash- 
 ing, &c. Vauquelain advised, in his ' Manuel de 1'Essayeur,' 
 to pour on the quartated cornet the weight of the assay 
 being 7*7 grains 554 to 770 grains of nitric acid (1*16 
 sp. gr.), which ought to fill the matrass half or two-thirds, 
 and boil gently for twenty, or twenty-two minutes at most, 
 to decant and replace the liquid by 500 to 800 grains of 
 acid (1-26 sp. gr.), and to boil for eight or ten minutes. The 
 assay is to be acted on always twice, because if we employ 
 at once very strong acid, the action will be too brisk, and 
 the cornet might be broken or carried out of the matrass, 
 and, on the other side, the acid of 1*16 sp. gr. cannot dis- 
 solve the last portions of silver, which are very difficult to 
 separate from the gold. 
 
 Surcharge. It is remarked that by following this 
 method the cornet always retains a small quantity of silver, 
 so that fine gold submitted to quartation and parting 
 always weighs more after than before the operation. The 
 augmentation of weight which it undergoes is termed the 
 surcharge: this surcharge is commonly from 0*001 to 
 0-002. M. Chaudet has found means to avoid it. In order 
 to do so, pour on to the quartated cornet nitric acid of 
 1*16 sp. gr., and heat for three or four minutes only ; 
 replace this acid by acid of 1'26 sp. gr., and boil during 
 ten minutes ; decant and make a second boiling with acid 
 of 1'26 sp. gr., which boil for eight or ten minutes. The 
 assay requires but from twenty to twenty-three minutes, 
 and, according to M. Chaudet, gives perfectly pure gold. 
 
 Mr. W. F. Lowe describes in the 'Chemica] News ' a use- 
 ful piece of apparatus for boiling gold assays. This appara- 
 tus consists of (1) a porcelain basin (a) 8^- ins. in diameter, 
 having a capacity of fifty oz. ; (2) a porcelain cover (b) 
 perforated with thirty holes, each hole being numbered in 
 black enamel; and (3) of a number of glass tubes (c). 
 These tubes are made to slip loosely through the holes in 
 the cover, and in order that they may not come against 
 the bottom of the basin the glass is bulged out into a ring 
 near the centre of the tube, which rests upon the cover ; 
 
764 
 
 THE ASSAY OF GOLD. 
 
 they have two small holes, one on each side of the bottom, 
 one in the centre of the bottom, and also one in the side 
 an inch above. 
 
 The method of employing the apparatus is the follow- 
 ing : Two basins containing a sufficient quantity respec- 
 tively of strong and weak nitric acid are heated over the 
 lamp, the weak acid basin being covered with the perfor- 
 ated cover carrying the boiling tubes, and the strong acid 
 
 FIG. 142. 
 
 (a) Section of basin, tubes, and cover, one-fourth size, (b) Cover, 
 one-fourth size, (c) Tube, half size. 
 
 'being covered with an ordinary dinner-plate. The assays, 
 which can be flattened and rolled while the acid is being 
 heated, are dipped into the tubes ; and in about five 
 minutes, if the acid is boiling, the whole thirty may be 
 lifted off by the cover, washed by being dipped into a 
 .basin of water, and then transferred to the basin of strong 
 .acid, where they are boiled for half an hour or more, after 
 
INQUARTATION. 765 
 
 which they are all lifted off by the cover, and transferred 
 to a similar basin full of water. Each tube is taken out 
 and plunged over head in water, so as to fill the tube, and 
 the assay is transferred to the crucible in the same way as 
 from a flask. 
 
 Mr. Lowe says that he has had this apparatus at work 
 for more than twelve months, and has found it a great 
 saving of labour, besides requiring so little attention ; in 
 fact, it can be left for an. hour without the assays taking 
 any harm, and he considers it preferable to boiling in? 
 flasks, for all the assays are under exactly the same con- 
 ditions, and can be boiled in the acid for a much longer 
 time. Another recommendation is that it is of very 
 moderate cost. For boiling the basins he uses two of 
 Fletcher's radial burners, which are very suitable, as they 
 are little affected by the fumes. 
 
 A source of loss occurs in parting operations and refin- 
 ing on the large scale, from the solution of gold in nitric 
 acid, even when it is quite free from hydrochloric acid, in 
 consequence of the formation of nitrous acid. To ascer- 
 tain the amount of loss from this source in ordinary assay 
 operations, Mr. Makin took four specimens of pure gold 
 accurately weighed, added the usual proportions of fine- 
 silver and lead, and then cupelled them. The resulting 
 buttons were rolled, coiled, and parted with nitric acid,, 
 the cornets being boiled in two acids of different strengths 
 a different number of times. Calling the weighings before^ 
 the operation 1000, the results were as follows : 
 
 1. Boiled in acid twice 999'6 
 
 2. three times 999-2 
 
 3. four 998-7 
 
 4. five ....... 997-9 
 
 The loss is thus seen to increase as the boilings are multi- 
 plied. 
 
 When silver is present in large quantity, Mr. Makin 
 believes that the solvent action of nitrous acid is restrained 
 by electrical action, the gold becoming the negative and 
 the silver the positive pole of a circuit ; but as the silver is 
 removed, the solution of the gold goes on more rapidly. 
 
766 THE ASSAY OF GOLD. 
 
 The cause of the evolution of nitrous acid is evident as long 
 as there is any silver present, and it often results from the 
 use of charcoal to prevent ' bumping.' When charcoal 
 is thoroughly carbonised, it does not materially affect the 
 acid ; but if it contain woody matter, nitrous acid is sure to 
 be set free. Mr. Makin has given up the use of charcoal 
 on this account. 
 
 The commercial importance of this subject will be 
 admitted, when we remember the enormous value of the 
 metals dealt with in this country, and that the question of 
 .profit and loss in commercial transactions with them are 
 -almost entirely in the hands of the assayer. A knowledge 
 of these facts may also serve to account for some of the 
 discrepancies between assayers. 
 
 In the assay of auriferous ores, the button produced 
 by cupellation commonly contains silver. When the pro- 
 portion of this metal surpasses that of inquartation, the 
 button is flattened between two pieces of paper, and treated 
 by pure nitric acid. The gold remains under the form of 
 a yellowish-brown powder, which is weighed immediately, 
 or fused in the cupel enveloped in a sheet of lead. 
 When the quantity is extremely small and imponderable, 
 we can assure ourselves at least of its presence by treat- 
 ing the residue left by nitric acid with aqua regia : if 
 it contain gold, it dissolves and gives a yellowish liquid, 
 in which a drop of solution of chloride of tin or the crys- 
 tallised chloride forms a deposit of purple of Cassius of 
 a violet colour : this character proves the presence of 
 the smallest traces of gold. When the gold predomin- 
 ates in the! bufton, o it is necessary to re-fuse it with three 
 times or less its weight of silver, and recommence the 
 assay. 
 
 Aqua Regia. When gold is the largest portion of the 
 alloy, and when there are reasons for not adding silver, 
 the parting can be made by aqua regia. In this case, all 
 the gold is dissolved, and the silver converted into chloride ; 
 the chloride is washed, dried perfectly, and weighed. When 
 the gold is precipitated by ferrous sulphate, it is washed 
 with a little hydrochloric acid, and annealed strongly 
 
SEPARATION OF GOLD AND SILVER. 767 
 
 before weighing, or even annealed so far as to fuse it, and 
 then cupelled with lead. 
 
 If an alloy, containing much silver, be treated by this 
 process, it sometimes happens that the excess of silver 
 chloride prevents the complete solution of the gold. In 
 this case it is necessary to reduce the alloy to an exces- 
 sively thin plate, to dissolve the chloride in ammonia, and 
 to treat afresh with aqua regia. This process can rarely 
 be made use of on the large scale, because the precipitation 
 of gold by ferrous sulphate is long and troublesome. 
 
 M. G. Eose fuses the alloy with lead, over a spirit- 
 lamp, in a porcelain crucible, acts on it with nitric acid, 
 which dissolves the silver and lead, precipitates the silver 
 by a solution of lead chloride ; lastly, the auriferous 
 residue is dissolved by aqua regia, and the gold precipitated 
 by ferrous chloride. 
 
 For the separation of gold and silver and their esti- 
 mation in alloys, H. von Jtiptner ('Zeitschrift ftir Analy- 
 tische Chemie,' 1879, 105) alloys the metal with 5 to 8 
 parts of zinc, and dissolves in nitric acid ; zinc, silver, 
 copper, &c., dissolve, whilst gold and the platinum metals 
 remain, and also tin as oxide, if present. The zinc alloy 
 is easily refined, and the metals in the crucible are best 
 covered with resin to prevent oxidation. 
 
 If it is known that neither tin nor platinum is present, 
 it is sufficient to decant, dry, and weigh in order to find the 
 weight of the gold. 
 
 If the platinum metals or tin are suspected the residue 
 is dissolved in aqua regia, the free chlorine is expelled by 
 boiling, the gold is reduced with ferrous ammonium sul- 
 phate of known strength, and the excess of ferrous oxide 
 titrated with potassium permanganate. From the quantity 
 of ferrous oxide consumed in the reduction of the gold 
 its proportion may be calculated. 
 
 All the silver is contained in the nitric acid solution of 
 the zinc alloy. 
 
 By another method for the separation of gold and silver 
 the metal is alloyed with 5 to 8 parts of zinc, for which 
 the heat of a Bunsen burner is sufficient. The alloy is 
 
768 
 
 THE ASSAY OF GOLD. 
 
 dissolved in nitric acid, when gold (with platinum and tin 
 as stannic oxide, if present) remains imdissolved. To- 
 separate gold from platinum and tin they are dissolved in 
 aqua regia, the platinum metals are precipitated with am- 
 monia, the free chlorine is expelled, and the liquid is mixed 
 with excess of ammonio-ferrous sulphate, the excess of 
 which is estimated by titrating back with permanganate 
 (< Zeitschrift fur Anal. Chemie,' 18, 104). 
 
 Standard of the Alloys of Gold. The real standard 
 of the alloys of gold is expressed in fractions of unity, as in 
 the case of alloys of silver. We suppose 24 carats in unity, 
 and 32-32nds in the carat ; the unity contains then 
 768-32nds. After these data the following table has been 
 formed, which expresses the relation of 32nds and carats to- 
 decimal fractions of the unity : 
 
 32nds 
 
 i 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 10 
 11 
 12 
 13 
 14 
 15 
 16 
 17 
 18 
 19 
 20 
 21 
 22 
 23 
 24 
 25 
 26 
 27 
 28 
 29 
 30 
 31 
 32 
 
 Decimals 
 
 Carats 
 
 0-001302 
 
 1 
 
 0-002604 
 
 2 
 
 0-003906 
 
 3 
 
 0-005208 
 
 4 
 
 0-006510 
 
 5 
 
 0-007912 
 
 6 
 
 0-009115 
 
 7 
 
 0-010415 
 
 8 
 
 0-011718 
 
 9 
 
 0-013021 
 
 10 
 
 0-014323 
 
 11 
 
 0-015625 
 
 12 
 
 0-016927 
 
 13 
 
 0-018230 
 
 14 
 
 0-019531 
 
 15 
 
 0-020833 
 
 16 
 
 0-022135 
 
 17 
 
 0-023436 
 
 18 
 
 0-024740 
 
 19 
 
 0-026042 
 
 20 
 
 0-027343 
 
 21 
 
 0-028644 
 
 22 
 
 0-029948 
 
 23 
 
 0-031250 
 
 24 
 
 0-032552 
 
 
 0-033854 
 
 
 0-035156 
 
 
 0-036460 
 
 
 037760 
 
 
 039062 
 
 
 0-040364 
 
 
 0-041667 
 
 
 Decimals 
 
 0-041667 
 0-083334 
 0-125001 
 0-166667 
 0-208333 
 0-250000 
 0-291666 
 0-333333 
 0-374999 
 0-416667 
 0-458630 
 0-500000 
 0-541667 
 0-583333 
 0-624555 
 0-666667 
 0-707333 
 0-750000 
 0-791666 
 0-833333 
 0-874999 
 0-916666 
 0-958333 
 1-000000 
 
ASSAY OP ALLOYS (ASSAY PKOPER). 
 
 
 
 Assays of Gold Coin and Bullion (Alloys of Gold and 
 Copper, or Gold, Silver, and Copper). 
 
 Preliminary Assay. As in the case of silver assaying 
 the quantity of lead to be employed is of importance, a 
 preliminary assay must be made when the standard of the 
 alloy to be examined is not approximatively known. It is 
 thus effected : To 2 grains of the alloy add 6 grains of 
 fine silver and 50 grains of pure lead. The lead must be 
 introduced into a hot cupel, and when fused, and its sur- 
 face fully uncovered, the alloy and silver may be added, 
 wrapped either in thin paper or a small quantity of lead 
 foil. The cupellation finished, and the cupel cold, the 
 button of gold and silver must be removed from the cupel 
 by aid of the pliers, and if necessary cleansed. Hammer it 
 to a thin plate on the anvil, place it in a small evaporating 
 basin, and treat it with half an ounce of nitric acid. (It 
 may be here mentioned, that the nitric acid employed in the 
 assay of gold must be chemically pure, and special care 
 must be taken that it contains no trace of chlorine.) The 
 evaporating basin is gently heated until all action ceases. 
 The brownish residue is repeatedly washed with hot water, 
 dried, ignited, and weighed ; and from its weight the 
 amount of lead and silver to be added in the actual assay 
 may be estimated. The presence of copper in the 'alloy 
 is indicated by the blackness of the cupel where it. is satu- 
 rated with oxide. 
 
 Assay Proper. In this case it will be supposed that 
 standard gold is the alloy operated on, and that prelimi- 
 nary assay has given about 91^ per cent, of gold. On 
 referring to the table (page 759), it will be found that 
 between 27 and 30 parts of lead are required for such per- 
 centage of gold, and that, according to the general observa- 
 tions on this class of assay, three times its weight (that is, 
 the weight of fine silver) will be required to so dilute the 
 gold that nitric acid can attack and dissolve out the whole 
 of the silver combined with it. 
 
 Place the weight representing 24 carats in the pan of 
 
 3D 
 
770 THE ASSAY OF GOLD. 
 
 the balance, and exactly counterpoise it with the gold to be 
 assayed ; two portions should be thus weighed. Two por- 
 tions of fine silver must now be weighed ; 33 grains will be 
 required for each 24 carats of gold, as 22 carats, or 11 
 grains, of fine gold exist in the 24 carats, and three times 
 the quantity of silver is necessary. 300 grains of lead must 
 be placed in a hot cupel (two being thus prepared), and, as 
 in the preliminary assay, when the surface is fully uncovered, 
 the gold and silver are added, and the cupellation proceeded 
 with, taking all the precautions already fully pointed out 
 elsewhere. 
 
 The button so obtained is cleaned, hammered on the 
 anvil, then annealed and passed between the rollers of a 
 small flatting-mill ; being occasionally annealed, in order to 
 prevent the laminated button cracking at the edges. When 
 reduced to the desired degree of thinness it is again an- 
 nealed and rolled round a quill or glass rod into a spiral, 
 termed a cornet. This cornet is placed in a parting flask 
 with 1^ oz. of nitric acid, sp. gr. 1'16, very gently heated 
 to the boiling point, and at that maintained for ten minutes. 
 The acid is then to be poured off, and 2 oz. of nitric acid, 
 sp. gr. 1*26, added, and again boiled for ten minutes. This 
 second acid is also poured off,' and a third quantity of like 
 specific gravity added and boiled. The cornet is then well 
 washed with distilled water, and the flask, filled with dis- 
 tilled water, is inverted, having its mouth closed with the 
 thumb. . The cornet will fall through the water without 
 breaking, and can be introduced, together with some of the 
 water, into a small crucible (cornet crucible), the water 
 poured off, the crucible and gold gradually dried, and then 
 heated to redness. When cold, the final operation of 
 weighing may be performed, thus : The weight represent- 
 ing 22 carats is placed in one pan of the balance, and the 
 cornet in the other : as the gold employed was supposed to 
 be standard, it ought to weigh exactly 22 carats. If, how- 
 ever, gold of greater or less fineness had been submitted to 
 assay say of 23 and 21 carats respectively 1 carat weight 
 would have been required in the pan containing the 22- 
 carat weight, to counterbalance the gold carat ; in this case 
 
ASSAY OF ALLOYS (ASSAY PKOPEE,). 771 
 
 the gold would be 23 carats fine, or, in the usual mode of 
 reporting, ' one carat better.' If, on the other hand, the 1 
 carat weight had been found necessary in the pan containing 
 the cornet, the gold would be 21 carats fine, or ' one carat 
 worse.' 
 
 In cases where it is known that the gold under exami- 
 nation contains no silver, the only alloy being copper, its 
 fineness can be estimated by cupelling 24 carats with its 
 proper portion of lead, and weighing the resulting button, 
 which should represent the amount of fine gold in the alloy 
 assayed. 
 
 Parting Assays. Parting assays are those assays by 
 which the amount of fine gold and fine silver in any alloy is 
 estimated. When the amount of gold exceeds that of the 
 silver, it is called ' gold parting ' ; when the amount of 
 silver exceeds that of the gold, ' silver parting.' 
 
 In this assay the weights employed in the silver assay 
 are employed, as the report is made in ounces of fine metal 
 per pound Troy. 
 
 12 grains (representing 1 Ib. Troy) of the alloy are 
 weighed off, cupelled with 300 grains of lead, and the 
 resulting button, containing only gold and silver, is weighed. 
 Suppose it weigh 10 grains, then 2 grains =2 ounces in 
 the pound of alloy, is copper or some other metal, which 
 has been oxidised and carried into the cupel with the 
 litharge. A preliminary assay must be made of the alloy, 
 to ascertain the approximative quantity of silver and gold, 
 so as to apportion the amount of silver in the assay proper : 
 this amount being found, it is to be weighed off, added to 
 the button of fine gold and silver obtained as above, and 
 the whole cupelled with 200 grains of lead ; the cupelled 
 mass of gold and silver laminated and treated with nitric 
 acid, as already described, and the resulting gold weighed. 
 Suppose the weight to be 8 grains =8 ounces, the result 
 would stand thus : 
 
 Copper or other base metal . . . 2 oz. 
 
 Gold 8 
 
 Silver 2 
 
 12 oz. 
 
 3 D 2 
 
77'2 
 
 THE ASSAY OF GOLD. 
 
 The above arrangement is very convenient for accom- 
 plishing gold assays, and is the one employed in the assay 
 office of the French Mint. The annexed cut (fig. 143) 
 represents this apparatus. 
 
 The assay flask, M 9 being charged with the cornet, a 
 constant amount of acid is added with a pipette. On the 
 addition of the second acid a small piece of charcoal is 
 
 FIG. 143. 
 
 placed in the flask ; this serves to prevent bumping during 
 ebullition. The flasks are supported on a plate of sheet 
 iron, P, pierced with holes or by a grating, and the acid 
 vapours, before escaping by the flue, pass into glass tubes, 
 JT, about half an inch in diameter, and four feet long : at 
 each end a narrower tube, , is fused. The lower tube 
 freely enters the neck of the flask ; and as the space 
 
 
ASSAY OF PYRITES FOR GOLD. 77a 
 
 between is so small that a layer of acid remains suspended 
 and obstructs the passage of the acid vapours, they are 
 thus forced to pass into the large tube, where, for the 
 greater part, they condense and fall into the flasks. By 
 this means the quantity of acid employed in the assay can 
 be diminished, as there is no loss by evaporation, and the 
 results are found to be more constant. In order that the 
 passage to the large tube for the acid vapours may always 
 remain free, the end of the narrow tube passing into the 
 flask must be cut at an angle (see P). The drops of acid 
 collect at this part, and never close the tube. 
 
 Assay of Pyrites for Gold. 
 
 Mr. J. M. Merrick gives the following method for 
 assaying pyrites for gold : 
 
 ' One pound, or even 18 ounces (avoirdupois), of fine 
 marble-dust is mixed with 8 ounces of finely pulverised 
 and sifted pyrites ; the whole then re-sifted and put into 
 a Hessian crucible, which should be about one-third filled 
 by the mixture. The crucible is set as usual on a fire- 
 brick, and a fire of hard coal is made around it, the coals 
 being heaped up to within an inch of the top. The cm* 
 cible is covered with a piece of brick or a piece of sheet 
 iron. During the first half-hour the contents should be 
 stirred once or twice. As the fire grows brisker the car- 
 bonic acid evolved keeps the contents of the crucible in 
 brisk ebullition, and the mixture should be stirred well 
 every five or ten minutes. On stirring during this time, the 
 iron rod used seems to meet with but little resistance 
 from the light mass, but at the end of about 1^ hour the 
 evolution of gas suddenly ceases, the red-hot mass becomes 
 heavy, sinks, and requires considerable force to keep it 
 stirred. It must be stirred well and vigorously, however, 
 for about half an hour, not leaving it unstirred for more 
 than a minute, otherwise the mass will fuse or cake, and 
 the assay will be almost inevitably ruined. 
 
 ' When a sample taken out in an iron spoon gives off 
 no smell of sulphur, the entire contents of the crucible 
 
774 THE ASSAY OF GOLD. 
 
 must be turned into a stoneware pot or a wooden bucket 
 half filled with water, and well stirred. When the powder 
 which should be uniform and free from lumps or fused 
 pieces has settled, the water must be poured off, the 
 wet mass allowed to drain, and then transferred to a large 
 earthen bowl or porcelain mortar. Here it is to be amal- 
 gamated with about 2 ounces of mercury, to which a little 
 bit of sodium amalgam has been added. The amalgama- 
 tion, as well as the stirring in the fire, is a tedious process, 
 and one which it is as well to do by proxy. It does not 
 consist in merely grinding with a pestle the mercury in 
 among the particles of the roasted ore, but this ore itself 
 must be ground in contact with the mercury, until the 
 particles are so fine that they will float suspended in water 
 for several seconds. At the end of say 10 minutes' 
 thorough grinding, the contents of the bowl are to be 
 brought into one mass in the bottom of the vessel, the 
 bowl then sunk in a tub of water, and the contents 
 * washed down ' an operation not easily described, but 
 familiar enough to every old miner. It consists essen- 
 tially in shaking the bowl half full of ore and water 
 in such a way that the mercury, gold particles, and un- 
 ground ore sink to the bottom, while the light and 
 finely ground ore is floated off into the tub. The ore re- 
 maining is re-ground and re-washed, and these processes 
 are repeated till nothing but the mercury remains in the 
 bottom of the bowl or mortar. This mercury is then 
 dried with filter-paper, and heated in a porcelain capsule 
 over a Bunsen flame, very gently, until it is sublimed and 
 the gold remains behind. The film of gold may then be 
 scraped up and melted, with a little borax and nitre, in 
 the very smallest-sized Hessian crucible, either with the 
 foot blowpipe or in a charcoal furnace, by which means a 
 round, clean button of gold, suitable for weighing, will be 
 obtained. 
 
 ' This method has its disadvantages and its counter- 
 balancing merits. On the one hand, it must be admitted 
 to be tedious, laborious, and to a considerable degree 
 uncertain. Some analysts fail with it altogether, while 
 
ASSAY OF PYRITES FOR GOLD. 775 
 
 none who have tried it, so far as I know, get closely 
 agreeing results. 
 
 ' But, on the other side, it is as certain that this 
 method will indicate the presence of gold, and will bring 
 out the gold in a weighable form from pyritic ores, where 
 the assay by smelting will not show a remote trace of the 
 precious metal ; and that where the fire assay shows a 
 certain percentage this will invariably bring out a larger 
 -amount. Large returns have been obtained by this 
 amalgamation method from iron pyritic ores, which have 
 been repeatedly assayed in the ordinary way, by chemists 
 of great eminence, with uniformly negative results.' 
 
 Treatment of Gold- and Silver-Bearing Copper Ores. 
 A very few words may serve to indicate the present prac- 
 tice in the separation of the precious metals from copper. 
 The older processes employed for this purpose were by 
 far the most complicated and wasteful operations known 
 to metallurgy, and it is only since the discovery and intro- 
 duction of the various ' wet ' processes that any but the 
 richest coppers could be advantageously treated for the 
 precious metals. 
 
 The Ziervogel process has only been successful in a 
 few isolated cases, and demands such pure material, and 
 such skill in manipulation, as to debar its use in ordinary 
 instances ; nor does it provide for the extraction of gold. 
 
 It is indisputable that the electrolytic methods are 
 rapidly advancing to the front in the treatment of gold- 
 and silver-bearing metallic copper, and have the great 
 advantages of producing a copper of the best quality, 
 but are yet largely in the experimental stage, and require 
 a bulky and expensive plant. 
 
 The' new Hunt & Douglas method, as applied to copper 
 ores or mattes, seems to fill the gap more completely 
 than any previous invention. By this method the copper 
 is extracted from the ore or matte after a very imperfect 
 roasting, and, being precipitated as a dioxide by sulphurous 
 acid generated from pyrites, it is decomposed by about 
 one-half its weight of metallic iron, the resulting cement 
 being fit for immediate refining. The copper is obtained 
 
776 
 
 THE ASSAY OP GOLD. 
 
 in a state of absolute purity even in the presence of ar- 
 senic and antimony ; while the residues, containing every 
 trace of the gold, silver, and lead originally present, may 
 be smelted with lead ores in a blast furnace. The pro- 
 cess has long passed the experimental stage, and offers 
 advantages peculiar to itself and unshared by any other. 
 
 The ease with which the small amount of gold some- 
 times present in cupriferous pyrites may be won is not 
 realised by all copper smelters, although the method is 
 extensively practised in this country, as well as at Swansea 
 and in Chili. 
 
 Owing to its great affinity for metallic copper, the gold 
 contained in white metal may be concentrated into a very 
 small bulk of the former by exposing the pigs of matte 
 to a slow oxidising fusion, exactly as in the process 
 for making blister copper. The operation, however, is in- 
 terrupted as soon as a certain quantity of metallic copper 
 is formed, when the furnace is tapped, and the product 
 now advanced to pimple metal, or even regulus, from 82 to- 
 88 per cent. being examined, bottoms of metallic copper 
 will be found under the first few pigs. This is the method 
 pursued in making best selected copper, for not only does 
 the small quantity of metallic copper extract the gold,, 
 but also the greater part of other foreign and injurious 
 substances, such as arsenic, antimony, tellurium, tin, &c.. 
 The proportion of bottoms formed must vary with the 
 quantity of gold present ; in some instances, even a repe- 
 tition of the processes being required to fully extract the 
 more valuable metal. Silver is but slightly concentrated 
 by this operation, as will be observed from the following 
 assays made under the author's direction : 
 
 Assay of Original 
 White Metal 
 
 Propor- 
 tion of 
 Bottoms 
 
 Assay of Bottoms 
 
 Proportion 
 thus 
 Extracted 
 
 Assay 
 of Residual 
 Pimple 
 
 Metal 
 
 Gold 
 
 Silver 
 
 
 Gold 
 
 Silver 
 
 Gold 
 
 Silver 
 
 Gold 
 
 Silver 
 
 Ounces 
 
 Ounces 
 
 Per cent. 
 
 Ounces 
 
 Ounces 
 
 Per cent. 
 
 Per cent. 
 
 Ounces 
 
 Ounces 
 
 0-64 
 
 93-3 
 
 6-4 
 
 9-60 
 
 213-4 
 
 93-7 
 
 14-8 
 
 0-030 
 
 78-7 
 
 2-37 
 
 16-6 
 
 9-0 
 
 19-10 
 
 36-2 
 
 90-2 
 
 18-5 
 
 0-110 
 
 14-2 
 
 0-11 
 
 
 
 5-4 
 
 1-73 
 
 
 
 88-4 
 
 - 
 
 0-012 
 
 
 
DETECTION OF MINUTE TRACES OF GOLD IN MINERALS. 777 
 
 In examining this table it must be remembered that a 
 considerable concentration has taken place in the matte 
 itself, as well as in the copper bottoms, so that the results 
 do not seem to agree ; but the figures given are sufficient 
 to indicate the general results of the process. Unless the 
 furnace bottom is already well saturated with auriferous 
 metal, a heavy loss in gold must be expected. 
 
 Detection of Minute Traces of Gold in Minerals. Mr. 
 Skey, analyst to the Geological Survey of New Zealand, 
 has devised a plan which gives very good results, even 
 when small quantities of mineral are operated on. He 
 employs iodine or bromine for the purpose of dissolving 
 out the gold. Both of these substances differ from chlo- 
 rine, especially in their relatively feeble affinities for hydro- 
 gen, so that there is less fear that from the generation 
 of hydrogen acids any great preponderance of other 
 matters would be dissolved along with the gold. Either 
 of these substances can be safely and advantageously 
 employed for the separation of gold from its matrix. 
 
 The following particulars of experiments made by this 
 method will be useful in showing what is approximately 
 the smallest quantity of gold that can be positively sepa- 
 rated and identified when operating upon a limited 
 quantity. 
 
 1st. 2 grms. of roasted ' buddle headings ' from a 
 quartz mine at the Thames, N.Z., known to contain gold 
 at the rate of 1 oz. or so to the ton, was well shaken for a 
 little while with its volume of alcoholic solution of iodine, 
 then allowed to subside.' A piece of Swedish filter-paper 
 was then saturated with the clear supernatant liquid, and 
 afterwards burned to an ash ; the ash, in the place of 
 being white, as it would be if pure, was coloured purple ; 
 the colouring matter was quickly removed by bromine 
 a clear indication of the presence of gold. The time 
 occupied by the whole process was twenty minutes. 
 
 2nd. 1 grm. of the same { buddle headings,' mixed 
 with such a quantity of earth as to reduce the proportion 
 of gold present to 2 dwts. per ton, was kept in contact 
 with its own volume of the tincture of iodine for two 
 
778 THE ASSAY OF GOLD. 
 
 hours, with occasional stirring ; a piece of filter-paper was 
 then saturated with the liquid, and dried, five times con- 
 secutively, and finally burnt off as before : in this case, 
 also, the colour of the residual ash was purple, and it 
 gave the reaction of gold. 
 
 3rd. 32 grms. of siliceous hematite, finely pounded, 
 were thoroughly mixed with precipitated gold to the 
 amount of % dwts. per ton ; then ignited and treated with 
 bromine water. After two hours the solution was filtered, 
 and evaporated to a bulk of 20 minims ; this gave a good 
 reaction of gold to the ' tin chloride ' test. 
 
 4th. 100 grms. of the hematite, with precipitated gold 
 at the rate of -J dwt. per ton, treated as before, but this 
 time well washed at the expiration of two hours ; the 
 washings evaporated along with the first filtrate gave 
 a fainter, but still decided, reaction of gold to the same 
 test. 
 
 5th. Iodine, as tincture, substituted for bromine in 
 Experiments 3 and 4, gave similar results ; the only varia- 
 tion made was, that, as a precautionary measure allowing 
 for its slower action, they were kept in contact for twelve 
 hours. 
 
 Careful experiments have been made to compare the 
 results of the common amalgamating process with the 
 foregoing, and it has been found that it is not certain, 
 with the same expenditure of labour, to get reliable indi- 
 cations of gold, when present in less quantity than 2 dwts 
 per ton, operating upon about 100 grms. of material. 
 
 In summing up the results of these experiments, it 
 appears, then, that for qualitative examinations for gold, 
 or for quantitative estimations in certain cases, iodine 
 and bromine are each superior to mercury. It also appears 
 that a proportion of gold equal to \ dwt. per ton, upon a 
 bulk of about 4 oz. of ferruginous matters, can be easily 
 and rapidly detected. Of course, by operating upon 
 larger bulks, gold could be discovered by this process, 
 were it present in far less quantities, but this is sufficiently 
 near for the majority of cases. 
 
 These processes are especially adapted for the sepa- 
 
DETECTION OF GOLD IN MINERALS. . 779 
 
 ration of gold from sulphides, as the preliminary roasting 
 is extremely favourable to them, the loss in the substitu- 
 tion of oxygen for sulphur amounting to 25 per cent, by 
 weight, while the volume remains constant (or nearly so) ; 
 hence there is a corresponding porosity in the product, 
 by which every particle of it is thrown open to contact 
 with the solution. This mechanical accessibility obviously 
 cannot be taken advantage of by mercury. 
 
 With sulphides these processes are practically ex- 
 haustive, while at the same time the simultaneous extrac- 
 tion of other matters is so trifling, that the proper tests 
 for gold can be safely applied directly to the concentrated 
 solution. In the roasting of pyrites it is necessary to 
 raise the temperature towards the end to a full red heat, 
 in order to decompose the ferruginous sulphates, since if 
 these remained iron would get into the solution. In the 
 case of an excess of calcium carbonate being present, it is 
 proper to gently reignite the roasted mineral, &c., with 
 ammonium carbonate, or much lime might get into the 
 iodine or bromine solution. On the other hand, a very 
 high temperature is to be avoided, for a considerable 
 quantity of fine gold can escape detection in this way 
 by the partial vitrification of the more fusible of the 
 silicates. 
 
 The identification of gold by the combustion of its 
 salts with filter- paper seems to promise a rapid method of 
 estimating it, comparatively, by the aid of a series of pre- 
 pared test-papers, representing gold in different degrees of 
 dilution. 
 
 Assay by the Spectroscope. It seemed at one time 
 possible that the assay of gold and silver alloys might be 
 simply and rapidly effected by the aid of the spectroscope. 
 The researches of Professor Chandler Eoberts, F.K.S., 
 Chemist to the Mint, and of Mr. A. E. Outerbridge, As- 
 sistant in the Assay Department of the United States Mint, 
 show that for the present at least these expectations are 
 groundless. 
 
 It has been shown by Mr. Capel that the ^Vo f a 
 milligramme of gold will show a spectrum, if the spark be 
 
780 THE ASSAY OF GOLD. 
 
 passed through a weak solution of the pure metal. But 
 when operating on a slip of alloy formed of 
 
 Silver 708 
 
 Copper 254 
 
 Gold 38 
 
 1,000 
 
 the spectra of copper and silver alone were visible. In 
 an alloy of gold and copper containing from 200 to 250 
 parts in the thousand of the precious metal, the gold 
 spectrum is barely visible. On the other hand, in an 
 alloy of gold and copper containing 1 per cent, of the 
 latter, the copper spectrum was distinctly shown. In 
 copper alloyed with 20 per cent, of nickel, the spectrum 
 of the latter is not visible. Hence we arrive at the in- 
 teresting fact that when two or more metals are present, 
 the spark will to some extent elect for its vehicle the one 
 which is most rapidly volatilised. 
 
 It is also not possible to obtain alloys of gold so per- 
 fectly homogeneous that the quantity of metal volatilised 
 and giving the spectrum may safely represent the whole 
 melt. 
 
781 
 
 CHAPTER XVIII. 
 
 THE ASSAY OF PLATINUM. 
 
 PLATINUM is found in a native or metallic state. It occurs 
 very rarely, yet it is exceedingly probable that wherever 
 gold is found this metal will more or less accompany it. 
 
 It is found disseminated in sand, in the form of grains 
 varying in size from gunpowder to hempseed : this last 
 size they rarely exceed, yet, as in the case of gold, nuggets 
 have been found of large size and weight. Its colour is 
 steel-grey, or, rather, a tinge between silver- white and steel- 
 grey. 
 
 The sands from which platinum is derived are remark- 
 able, from the number and importance of their principal 
 constituents. With the platinum may be found Au, Ag, 
 Hg, Fe, Cu, Cr, Ti, Ir, Os, Eh, Eu, and Pd. Besides all 
 these metals, precious stones have also been found as- 
 sociated with it. 
 
 The following plan will serve to detect platinum in 
 admixture with gold and other heavy matters obtained by 
 washing or vanning sands, earths, &c. 
 
 Act on a small quantity by mercury, and separate the 
 amalgam ; by this means the gold is removed. To the 
 residue add aqua regia and boil, evaporate the solution to 
 dryness, add a little hydrochloric acid and water, boil and 
 filter. To the filtered solution add a strong solution of sal 
 ammoniac (ammonium chloride). If a bright yellow, or 
 reddish-yellow, granular precipitate falls, platinum is pre- 
 sent in the sand. 
 
 A still more ready method is the following : Separate 
 as much earthy matter as possible by careful washing. If 
 gold is present, separate that by amalgamation. Dry the 
 
782 
 
 THE ASSAY OF PLATINUM. 
 
 residue and take its specific gravity ; if it be above 10, 
 platinum is most likely present. The specific gravity of 
 native platinum, free from earthy matter, is from 16 to 19. 
 
 Analysis of Platinum Ores. The platinum sands often 
 contain metallic compounds of iron and platinum, not only 
 capable of being attracted by the magnet, but possessed 
 even of polarity. These grains have a different composition 
 from those not magnetic, as shown in the two following 
 analyses by Berzelius : 
 
 Analysis of the non-magnetic grains : 
 
 Platinum 78-94 
 
 Iridium 
 
 
 
 
 
 
 4-97 
 
 Rhodium 
 
 
 
 
 
 
 86 
 
 Palladium 
 
 
 
 
 
 
 28 
 
 Iron . 
 
 
 
 
 
 
 11-04 
 
 Copper 
 
 
 
 
 
 
 70 
 
 C ]Q 2fI*tlIlS 
 
 
 
 1-00 
 
 LUm \in scales 
 
 
 
 96 
 
 Analysis of the magnetic grains : 
 
 Platinum 
 
 Iridium 
 
 Rhodium 
 
 Palladium 
 
 Iron '.' 
 
 Copper 
 
 Insoluble matters 
 
 98-75 
 
 73-58 
 
 2-35 
 
 1-15 
 
 30 
 
 12-98 
 5-20 
 2-30 
 
 97-86 
 
 These grains being separated, their relative proportion 
 is estimated. 
 
 Bunsen's method of analysing platinum ores is as fol- 
 lows : The ores employed contain no osmium, and were 
 relatively rich in rhodium. 
 
 Platinum, and Palladium. It is easy to effect the 
 almost complete separation of platinum and palladium 
 from rhodium, iridium, and ruthenium. The original 
 material is mixed in a Hessian crucible, with from ^ to J- 
 its weight of ammonium chloride, heated until the latter 
 is completely volatilised, allowed to glow gently until only 
 the vapours of ferric chloride show themselves, and then 
 placed in a porcelain dish, with from two to three times 
 
ASSAY OF PLATINUM ORES. 783 
 
 its weight of raw commercial nitric acid, and evaporated 
 to a syrupy consistency. By this treatment with am- 
 monium chloride the metals present not belonging to the 
 platinum group will have been partially converted to lower 
 chlorides, the rhodium, iridium, and ruthenium will have 
 been rendered insoluble, and the silica present as gangue 
 converted from a gelatinous mass to a finely pulverulent 
 condition, in which state it will admit of speedy filtering. 
 
 The chlorine compounds, produced by the ammonium 
 chloride, give, upon digestion with nitric acid, just enough 
 hydrochloric acid to dissolve the platinum to bichloride, 
 while the metallic copper and iron present act so far re- 
 ducingly upon the palladium (in solution in nitric acid) 
 that it remains in solution, not as bichloride, but as the 
 protochloride, which latter is not precipitated with potas- 
 sium chloride. The mass is diluted with water, filtered, 
 and the solution saturated with potassium chloride, and 
 the greater part of the platinum separated pure as potas- 
 sium platinochloride, which is washed out first with potas- 
 sium chloride, and later with absolute alcohol (the last 
 washings must not be added to the solution). 
 
 The filtrate is poured into a large flask (which can be 
 made airtight), which will not be more than half-filled with 
 it. Chlorine gas is passed into this flask, and it is from 
 time to time shaken vigorously, until no further absorption 
 of gas takes place, when all the palladium will have sepa- 
 rated as a cinnabar-red precipitate of potassium palladio- 
 chloride (somewhat impure, however, from traces of 
 platinum, iridium, and rhodium). The fluid from which 
 these precipitates were obtained is now evaporated, not 
 quite to dryness, with hydrochloric acid ; and, upon ad- 
 dition of just so much water as is necessary to dissolve out 
 the potassium chloride and other soluble salts (aiding the 
 operation by rubbing with a pestle), there remains behind 
 a dirty, yellow-coloured precipitate. This is separated by 
 filtration, boiled with caustic soda and a few drops of 
 alsolute alcohol. Hydrochloric acid is added to dissolve 
 the precipitate formed, and the liquid then saturated with 
 potassium chloride ; the result is a precipitate of chemically 
 
784 THE ASSAY OP PLATINUM. 
 
 pure potassium platinochloride. The mother-liquid con- 
 tains only copper and no platinum metals. 
 
 The purification of the cinnabar-red precipitate of pal- 
 ladium is accomplished as follows : Dissolve in boiling 
 water, whereby a portion of the chloride dissolves, with 
 evolution of chlorine, to palladium protochloride. Then 
 evaporate with 2^ times its weight of oxalic acid, and dis- 
 solve again in a solution of potassium chloride ; where- 
 upon potassium platinochloride remains behind, chemically 
 pure. Wash out as before. 
 
 The brown liquid is then somewhat concentrated upon 
 the water-bath : and upon cooling, there separate bright 
 green, well-formed crystals of potassium palladio-proto- 
 chloride (with some potassium chloride), which upon 
 testing proves free from the other platinum metals. 
 
 The fluid poured off from these crystals is then 
 neutralised carefully with caustic soda, and gives a slight 
 precipitate of copper and iron, which is filtered off. 
 Upon adding potassium iodide to the filtrate, all the 
 palladium separates as palladium iodide. To avoid adding 
 an excess of the reagent, it is best to take from time to 
 time a drop from the fluid with a capillary tube, and put 
 the same upon a watch-glass. As long as the precipitation 
 is incomplete, the drop appears, upon a white background 
 brown ; when complete, it is colourless ; when the reagent 
 is present in excess it is red. This is tested for its purity 
 by reducing it to metallic palladium, and then heating and 
 dissolving in nitric acid ; when pure, it must dissolve 
 completely. The whole mass is now reduced in a slow 
 stream of hydrogen gas (whereby the iodine can be 
 obtained again, as hydriodic acid, by absorbing with 
 water). At last the mass must be strongly heated, to 
 decompose slight traces of the palladium subiodide which 
 are formed. 
 
 The mother-liquid from which all this platinum and 
 palladium have been obtained may contain some iridium 
 and rhodium ; it is, therefore, evaporated to dryness with 
 a little potassium iodide, whereby a mixture of rhodium 
 and iridium iodides separates. This can either be dissolved 
 
ASSAY OF PLATINUM OKES. 75 
 
 in aqua regia, and the two metals separated (as will here- 
 after be described) by sodium bisulphite, or it can be 
 united with the next portion from which these metals will 
 be obtained. 
 
 Ruthenium, Rhodium, and Tridium. The residue from 
 the original material which remains, after treatment with 
 ammonium chloride and nitric acid, is treated as follows, 
 to get the metals in a form adapted to further chemical 
 treatment. 
 
 The method depends upon the behaviour of zinc 
 chloride to zinc. If a piece of zinc be melted, it rapidly 
 covers itself with a stratum of oxide. If, to the melted 
 metal, a metal like iridium be added, the oxide stratum 
 hinders the latter from coming into contact with the zinc, 
 even though it be pushed beneath the surface. If, how- 
 ever, a few grains of ammonium chloride be added to it, 
 ammonia, hydrogen, and zinc chloride will be formed, 
 which last dissolves the oxide stratum to basic zinc chlo- 
 ride. The zinc below resembles mercury in lustre and 
 mobility. As soon as the chloride has dissolved as much 
 of the oxide as is possible for it, the oxide stratum again 
 forms, and is instantly removed again by the addition of 
 more ammonium chloride. The melted zinc, strewn with 
 ammonium chloride, also possesses, like mercury, the 
 property of attacking other metals, if the affinity exists 
 of forming alloys with them. By strewing ammonium 
 chloride upon the melted zinc, a quiet surging is kept 
 up, as the ammonia and hydrogen are given off. Many 
 oxides and chlorides (among which are those of the 
 platinum metals), when they come into contact with this 
 atmosphere of reducing gases, and with the basic zinc 
 chloride, are instantly reduced and dissolved to alloys by 
 the zinc. In making the solution, the zinc, in a porcelain 
 dish, should be constantly rotated : the gangue remains 
 in the basic chloride. The regulus, immediately upon 
 solidifying, should be taken from the capsule, out of the 
 yet fluid basic chloride, and washed off with acetic acid 
 until all the basic chloride is dissolved away. The gangue 
 can be quantitatively estimated by filtration and weighing. 
 
 3E 
 
786 THE ASSAY OF PLATINUM. 
 
 If the regulus is not immediately removed, the containing 
 vessel will be broken, owing to the unequal expansion of 
 the porcelain and the metal. 
 
 The best proportions for a quantitative separation 
 are, to 1 part of the platinum metals, from 20 to 30 parts 
 of zinc. For an ordinary separation 7 parts of zinc are 
 sufficient. 
 
 For the extraction of the residues remaining after the 
 treatment with nitric acid, this method is admirably 
 adapted. By fusing only once with zinc for two or three 
 hours, all the platinum metals are extracted. The opera- 
 tion is as follows : 
 
 From 3 to 3*5 kilos, of commercial zinc are fused in a 
 2-litre Hessian crucible, ammonium chloride from time 
 to time strewn upon it ; 400 grms. of residue, previously 
 heated to faint glowing with ammonium chloride, are 
 added, and the temperature kept, for two or three hours, 
 just above the fusing-point of the alloy, by adding, when- 
 ever the mass threatens to solidify, some ammonium 
 chloride. The mass is divided into three strata after 
 solidification has taken place. 
 
 The outer stratum, easily broken away by a blow from 
 a hammer, contains no platinum metals. The next con- 
 tains some particles of the zinc and platinum alloy, im- 
 bedded in the basic zinc chloride ; it is porous, and not 
 very thick. The inner stratum consists of a beautiful 
 crystalline regulus. 
 
 To obtain the alloy from the middle stratum, it is 
 only necessary to wash repeatedly with water ; and the 
 alloy gained is, of course, to be added to the regulus. To 
 obtain this regulus as pure as possible, it is again fused 
 with 500 grms. of zinc and some ammonium chloride, then 
 granulated in water, and the granules dissolved in fuming 
 hydrochloric acid. The acid attacks the regulus with 
 greatest energy, and the solution is complete in less than 
 an hour. The zinc chloride can be used for the next 
 operation. 
 
 The platinum metals are found at the bottom of the 
 vessel, in the form of a finely divided black powder, which 
 
ASSAY OF PLATINUM ORES. 787 
 
 is contaminated with zinc, and with traces of iron, copper, 
 &c., from the latter. It cannot be purified with nitric 
 acid, nor with aqua regia, for part of the platinum metals 
 will thereby be dissolved, or, at best, so suspended in the 
 fluid that filtration is impossible. If, however, the powder 
 is treated with hydrochloric acid, singularly enough, all 
 the impurities are dissolved ; not only zinc and iron, but 
 also lead and copper, dissolve readily with the generation 
 of hydrogen. The explanation is readily found in elec- 
 trical currents produced by the contact of the metals, the 
 stream passing from the positive zinc, iron, &c., to the 
 negative platinum metals, hydrogen being given from the 
 latter, and chlorine from the former, and uniting with 
 them. The metallic powder, after thorough washing, 
 possesses the property, upon being gently heated, of ex- 
 ploding weakly, and, when highly heated, with violence, 
 the explosion being accompanied with the evolution of 
 light, although neither hydrogen, chlorine, nitrogen, nor 
 aqueous vapour is given off; and as these are the only 
 elements which it is possible that the metallic powder 
 could have taken up, it must be assumed that these metals 
 are, by this treatment, converted into an allotropic con- 
 dition, and that, upon heating, they return, with more 
 or less energy, to their original condition. The powder 
 contains, mainly, rhodium and iridium ; but there are 
 traces present of platinum, palladium, lead, copper, iron, 
 and zinc. 
 
 It is intimately mixed with about 3 or 4 times its 
 weight of completely anhydrous barium chloride, and a 
 stream of chlorine gas led over it at a tolerably high 
 temperature. The operation is concluded when particles 
 of ferric chloride show themselves on the neck of the flasks 
 containing the powder. These are carefully brushed away 
 with filter-paper. Some water is now added, and the 
 mass of the platinum metals dissolves with the evolution 
 of heat. There remains behind insoluble matter, which, 
 upon reduction with hydrogen, alloying with zinc, and 
 treatment with hydrochloric acid, furnishes ruthenium. 
 From the solution all the barium chloride is removed by 
 
 3 E 2 
 
788 THE ASSAY OF PLATINUM. 
 
 careful addition of sulphuric acid. The platinum metals 
 are now completely freed from all other metals by reduc- 
 tion with hydrogen, the temperature being, throughout 
 the operation, maintained at nearly 100 C., by means of 
 a constant water-bath. Platinum and palladium chiefly 
 separate first ; then mainly rhodium ; and the last por- 
 tions consist almost entirely of iridium. It is best to 
 break off the operation when the fluid has assumed a 
 greenish-yellow colour. The last portions of iridium 
 (obtained by evaporating the solution to dryness, fusing 
 with sodium carbonate, and treatment with aqua regia) 
 are added to the portion, afterwards to be again rendered 
 workable by renewed treatment w r ith barium chloride. 
 The operation of reduction is hastened by concentrating 
 the fluid ; in doing which care must be taken to guard 
 against explosion, on account of the hydrogen. The 
 separated metals are treated with aqua regia, and the 
 platinum and palladium thus dissolved separated from 
 each other as already described. The traces of rhodium 
 and iridium in the mother-liquid can be removed entirely 
 by continued boiling with potassium iodide (whereby they 
 precipitate as iodides) ; they are then dissolved in aqua 
 regia and added to the insoluble portion. 
 
 This insoluble and partly oxidised portion is now again 
 reduced by hydrogen gas, treated as before described, with 
 barium chloride, and, after the removal of the barium, 
 the last traces of platinum and palladium are removed by 
 boiling with caustic soda. Ehodium and iridium now alone 
 remain to be separated. 
 
 The brown-red fluid is, for this purpose, evaporated 
 with hydrochloric acid, and, after filtration, treated with 
 sodium bisulphite in great excess, and the whole allowed 
 to remain quietly in the cold for several days. The double 
 rhodium and sodium sulphite separates slowly, giving a 
 lemon-yellow precipitate. The solution becomes lighter 
 and lighter, and finally almost colourless. The colour of 
 the precipitate changes with that of the fluid, becoming,, 
 with it, lighter. This precipitate, upon washing, contains 
 the rhodium almost pure. 
 
ASSAY OF PLATINUM ORES. 780 
 
 Upon heating the fluid gently, a yellow-white preci- 
 pitate separates, which consists mainly of rhodium, but 
 contains also some iridium. After filtering off this pre- 
 cipitate, the solution, upon being concentrated to a small 
 volume, gives yet two precipitates 
 
 1. A curdy, slowly separating, yellowish white pre- 
 cipitate containing nearly chemically pure iridium, with 
 but the faintest traces of rhodium. 
 
 2. A heavy crystalline powder, quickly separating, 
 which is readily freed from the first by decantation. Upon 
 testing, it gives all the reactions for iridium, but likewise 
 some peculiar reactions not shown by the latter. 
 
 The complete separation of rhodium from iridium is 
 accomplished by treating the yellow precipitates with 
 concentrated sulphuric acid. They are added in small 
 portions to the acid, heated in a porcelain capsule until 
 all the sulphurous acid has escaped, and then left upon the 
 sand-bath until the free sulphuric acid has been driven off 
 and sodium sulphate formed. Upon boiling the mass in 
 water, all the iridium dissolves as sulphate, with a chrome- 
 green colour, while the rhodium remains behind as a flesh- 
 coloured double salt of sodium and rhodium. The latter 
 is boiled in aqua regia, and washed by decantation. It 
 is insoluble in water, hydrochloric or nitric acids, and in 
 aqua regia. The rhodium and iridium are now completely 
 separated. 
 
 The first yellow precipitate obtained in the cold by the 
 sodium bisulphite gives, by this treatment, the rhodium 
 quite pure. The second and third precipitates, containing 
 much iridium, give very fine rhodium, but still slightly 
 contaminated with iridium. The products, therefore, 
 obtained by this treatment with sulphuric acid (which 
 betray their contamination with iridium by their somewhat 
 brownish colour) are collected for themselves, the rhodium 
 separated therefrom by glowing, treated again with barium 
 chloride, and the operation of separation repeated. The 
 green solution, containing only iridium, is gradually heated 
 over an ordinary burner, in a porcelain capsule, and after- 
 wards upon the sand-bath, to remove the excess of sul- 
 
790 THE ASSAY OF PLATINUM. 
 
 phuric acid ; and, finally, the capsule and its contents are 
 highly heated in a Hessian crucible. There is formed 
 thereby sodium sulphate and iridium sesquioxide. Upon 
 boiling the mass with water, the last remains behind as 
 a black, insoluble powder, which is readily washed by 
 decantation. 
 
 C. Lea's Process for Analysing Platinum Ores. The 
 ores on which these analyses were performed contained 
 chiefly iridium, together with ruthenium, osmium, rhodium,, 
 and platinum. It was a Californian osni-iridium which had 
 already undergone a preliminary fusion with nitre and 
 caustic potash. 
 
 This material is boiled with aqua regia to extract all 
 the soluble portions, the residue then ignited with nitre 
 and caustic soda,* and the fused mass heated with water. 
 From the resulting solution small portions of potassium 
 osmite crystallise out. The metallic oxides are next pre- 
 cipitated, and this precipitate, together with the portions 
 insoluble in water, is boiled again with aqua regia, ignited 
 again, &c. These ignitions still leave a small portion of 
 un at tacked residue. 
 
 The boiling with aqua regia is continued for a long 
 time, in order to get rid as thoroughly as possible of the 
 osmic acid. Even 200 hours' boiling, however, still leave 
 osmium in the solution in easily recognisable, but in com- 
 paratively small, quantity. The greatest advantage is 
 found throughout the whole of this part of the operation 
 from the use of a blowing-apparatus, by the aid of which 
 all inconvenience from the fumes of osmic acid is avoided. 
 The apparatus is constantly swept clear by a powerful air- 
 current, and the osmic acid is removed as fast as it is 
 volatilised. As the ignition of the ore with alkaline nitrate 
 and caustic alkali scarcely drives off any osmium, and 
 as almost all inconvenience in manipulating the resulting 
 
 * Attention is necessary to the order in which these substances are em- 
 ployed. If the caustic soda is melted first, it attacks the iron vessel strongly, 
 and may even go through. If added last, it causes sudden and violent effer- 
 vescence, with danger of boiling over. Therefore, place the nitre first in the 
 vessel, and when it is fused add the caustic soda. When a red heat is obtained 
 add the osm-iridium by degrees. 
 
c. LEA'S PROCESS. 791 
 
 solutions can be avoided by throwing down. the metals with 
 alcohol from the hot alkaline solution, in place of using 
 acid, it is clear that the difficulties arising from the noxious 
 effects of osmic acid can be almost wholly removed from 
 each of the various stages of the process. 
 
 A very prolonged treatment with aqua regia is found to 
 have the great advantage of converting nearly the whole 
 of the ruthenium into bichloride. The separation of 
 ruthenium in this form from the other metals is so easy in 
 comparison with the difficulties presented by the separation 
 of the sesquichloride, that this advantage cannot be looked 
 upon as other than a very material one. 
 
 Sal-ammoniac is next added to the mixed solution in 
 quantity sufficient to saturate it. The sandy crystalline pre- 
 cipitate (A) is thoroughly washed out, first with saturated, 
 and then with dilute sal-ammoniac solution. The saturated 
 solution of ammonium salt carries through with it nearly 
 the whole of the ruthenium as bichloride (B) ; the dilute 
 solution is found to contain small quantities of iridium, 
 rhodium, and ruthenium (C). 
 
 Over (A), water acidulated with hydrochloric acid is 
 placed, and allowed to stand for some days. This is treated 
 with ammonia and boiled. The precipitate, when treated 
 with hydrochloric acid, furnishes green osmium chloride, 
 with traces of ruthenium. 
 
 In these preliminary steps, Claus's process has been 
 followed, which undoubtedly offers advantages over any 
 other, and best brings the metals into a convenient state 
 for separation, varying it only by prolonging the treatment 
 with aqua regia^ and converting the ruthenium principally 
 into bichloride instead of sesquichloride. 
 
 We have now three portions of material: (A), con- 
 sisting of ammonium iridiochloride, containing also ruthe- 
 nium, osmium, rhodium, and platinum in small quantities. 
 (The ore under examination contained no palladium, which 
 metal, if present, has always its own peculiar mode of 
 separation, and does not increase the difficulties of opera- 
 tion.) (B), containing ruthenium bichloride, together 
 with iron in quantity, copper, and other base metals which 
 
792 THE ASSAY OF PLATINUM. 
 
 may be present. Finally (C), containing chiefly ruthenium 
 bichloride, mixed with small quantities of iridium and 
 rhodium. 
 
 The next step in the process is to introduce the am- 
 monium iridio-chloride (A) into a large flask with twenty 
 to twenty-five times its weight of water, and apply heat 
 until the solution is brought to the boiling-point ; the 
 whole of the ammonium iridio-chloride should be brought 
 into solution in order that the reduction to be effected 
 may not occupy too long a time, as otherwise the platinum 
 and ruthenium salt, if any be present, might likewise be 
 attacked. Crystals of oxalic acid are thrown in as soon 
 as the solution actually boils, whereupon a lively efferves- 
 cence takes place, and the iridium salt is rapidly reduced. 
 As fast as the effervescence subsides, more oxalic acid is 
 added until further additions cease to produce any effect. 
 When this is the case, the liquid is allowed to boil for two 
 or three minutes longer, not more ; the heat is to be re- 
 moved, and the flask plunged into cold water. 
 
 By this treatment any platinum present is unaffected. 
 Sal-ammoniac in crystals is added, about half enough to 
 saturate the quantity of water present. The sal-ammoniac 
 may be added immediately before the flask is removed 
 from the fire. After cooling, the solution should be left for 
 a few days in a shallow basin, whereby the ammonium 
 platino-chloride will separate out as a yellow, reddish, or 
 even (especially if the quantity of water used was insuf- 
 ficient) as a black crystalline powder, according to the 
 quantity of iridium which it may contain. 
 
 The mother-liquor is to be again placed in a flask and 
 boiled with aqua regia. On cooling, the ammonium iridio- 
 chloride crystallises out, and any traces of rhodium and 
 ruthenium which maybe present remain in solution. The 
 iridium salt is to be washed with a mixture of two parts 
 of a saturated solution of sal-ammoniac and three parts of 
 water, and may then be regarded as pure. 
 
 The treatment by oxalic acid affords iridium free from 
 all traces of ruthenium. 
 
 The treatment of solutions (B) and (C) presents no 
 
c. LEA'S PKOCESS. 793 
 
 difficulty. With (B) the best plan is to place the solution 
 aside in a beaker covered with filter-paper for some time. 
 Treated in this way, the bichloride gradually crystallises 
 out, and by re-crystallisations may be obtained in a state 
 of perfect purity. 
 
 Solution (C) is to be evaporated to dryness, and re- 
 duced to an impalpable powder. It is then to be thrown 
 upon a filter, and thoroughly washed with a perfectly 
 saturated solution of sal-ammoniac. The ruthenium bi- 
 chloride is thus carried through, with perhaps a trace of 
 rhodium sesquichloride, from which, however, it is easily 
 freed by crystallisation. From the residue, the rhodium 
 and ammonium sesquichloride is removed by a dilute solu- 
 tion of sal-ammoniac, perfectly free from the iridium, w r hich 
 is left behind. 
 
 In connection with this separation, Mr. Lea makes a 
 remark which, though of special reference to this par- 
 ticular case, is also applicable to all those cases in which 
 the double chlorides of the platinum metals are to be 
 separated by their various solubilities in solution of sal- 
 ammoniac. This most valuable process, for which we are 
 indebted, as for so much else, to Glaus, whose untiring 
 labours have made him the father of this department of 
 chemistry, requires to be applied with some attention to 
 minutiae. 
 
 The crystalline matter must be reduced to the finest 
 powder, and after being thrown upon the filter it must be 
 washed continuously until the separation is effected. Any 
 interruption of the washing is followed by more or less 
 crystallisation of sal-ammoniac through the material, which 
 precludes an effectual separation. The same material, which 
 in a state of coarse powder will hardly yield up enough 
 ruthenium bichloride to colour the sal-ammoniac solution, 
 will, when thoroughly pulverised, give an almost opaque 
 blood-red filtrate. 
 
 Solution (C) may be subjected to a different treatment 
 from the foregoing, and oxalic acid may be used to effect 
 the separation. The solution is to be brought to the boil- 
 ing-point, and oxalic acid added as long as effervescence 
 
794 THE ASSAY OF PLATIXUM. 
 
 is produced. The iridium bichloride is thereby reduced ; 
 the ruthenium bichloride and the rhodium sesquichloride 
 are not affected. Sal-ammoniac is then to be dissolved 
 in the solution to thorough saturation. By standing and 
 repose the double rhodium and ammonium chloride sepa- 
 rate out. The solution is then re-oxidised by boiling with 
 aqua regia ; by standing for some days in a cool place, the 
 ammonium iridio-chloride crystallises out, and the super- 
 natant solution contains the double ammonium chloride 
 and ruthenium bichloride, which may be rendered pure 
 by several re-crystallisations. 
 
 For purifying the double iridium and ammonium chlo- 
 ride the oxalic process is decidedly the best. It is simple 
 and less troublesome, and there is the further advantage 
 that the platinum is left in the condition of double chloride ; 
 whereas when the usual method of treating with aqueous 
 sulphuretted hydrogen is used, the platinum is apt to be 
 converted partly into sulphide, together with any traces 
 of rhodium and ruthenium which may be present. When 
 oxalic acid is used, the platinum remains behind as a red- 
 dish powder, containing some iridium, from which it may 
 be freed in the ordinary manner, if it is present in quantity 
 sufficient to be worth working. 
 
 For treating a mixture such as that which is here 
 designated as (C), containing no platinum, and only ruthe- 
 nium present in the form of ammonium ruthenio-chloride, 
 it is unnecessary to apply reducing-agents, and the first 
 method described is the best. But if it be proposed to 
 effect the separation by the reduction of the iridium com- 
 pound, the method here described is preferable to that 
 based on the use of sulphuretted hydrogen even in this 
 case. 
 
 The action of oxalic acid on the platinum metals is 
 interesting ; its reducing effect upon iridium bichloride at 
 the boiling-point is immediate. On ruthenium bichloride 
 it seems to have no effect whatever, and they may be boiled 
 together for a length of time without sensible result. In 
 a trial made with ruthenium and ammonium sesquichlo- 
 ride, the oxalic acid was boiled with the metallic salt for a. 
 
DEVILLE AND DEBRAY S PROCESS. 795 
 
 considerable time without any effect becoming visible, but 
 by long-continued boiling a gradual precipitation took 
 place. When ammonium platino-chloride was boiled with 
 oxalic acid, no effect was produced for a considerable 
 time, but gradually the platinum salt diminished in quan- 
 tity, and the liquid acquired a stronger yellow colour, 
 perhaps owing to formation of soluble platinic oxalate. 
 This process will not, however, furnish an easy and conve- 
 nient method of purifying commercial platinum from the 
 iridium always found in it, as the reduction of very small 
 quantities of double iridium and ammonium chloride in 
 the presence of a large proportion of the corresponding 
 platinum salt is difficult and slow, and the platinum salt 
 itself is evidently attacked. 
 
 The following method of Analysis of Platinum Ores, by 
 MM. Deville and H. Debray, may be useful. The ores of 
 platinum contain the following substances : 
 
 1. Sand. The whole of the sand is never removed 
 by washing the ore ; and the sand contains quartz, 
 zircon, chromate of iron, and, in the Russian ores, titanate 
 of iron. 
 
 2. Osm-iridium. 
 
 3. Platinum, iridium, rhodium, and palladium, com- 
 bined, no doubt, in the form of an alloy. 
 
 4. Copper and iron, which exist in the ores as an 
 alloy, for the iron found in the sand is not soluble in 
 acids. 
 
 5. Gold, and, oftener than is supposed, a little silver. 
 The latter metal is generally found with the palladium, 
 and it is very rarely that palladium is obtained quite free 
 from silver when it is prepared by the old processes. 
 
 1. Sand. To estimate the sand, we take a small assay- 
 crucible, or an ordinary crucible with smooth sides, and 
 melt in a little borax, so as to glaze the inside. We now 
 introduce from 7 to 10 grammes of pure granulated silver, 
 and 2 grammes of the ore fairly taken, and weighed very 
 accurately. Over the platinum we put 10 grammes of 
 fused borax, and one or two small pieces of wood charcoal. 
 The silver is now melted, and care must be taken to keep 
 
796 THE ASSAY OF PLATINUM. 
 
 it for some time a little hotter than the melting-point, so 
 that the borax may be very liquid, and may dissolve the 
 vitreous matters which accompany the platinum and con- 
 stitute the sand. The crucible is now allowed to cool, and 
 when it is cold the button, which will contain the silver, 
 osmium, platinum, and all the other metals, is detached, 
 and if necessary digested for a time with weak fluoric acid 
 to remove the last portions of borax. It is now heated to 
 a faint redness, and then weighed. The weight of the 
 button, subtracted from the sum of the weight of the ore 
 and silver employed, will give the amount of sand con- 
 tained in the ore. For example : 
 
 Milligr. 
 
 Californian ore ...... 2000 
 
 Silver ......... 7221 
 
 9221 
 After fusion, the button weighed . . 9162 
 
 Consequently, the ore contained, sand . 59 
 
 It is very important to know this number, for it represents 
 the only matter absolutely destitute of value which the 
 ore contains ; and this simple operation may be considered 
 the most important performed in estimating the value of 
 an ore. It is, besides, performed so quickly, that it is as 
 well to do at the same time two or three specimens, taken 
 from different parts of a lot of platinum powder. 
 
 2. Osm-iridium. Another 2 grammes of the ore weighed 
 very accurately are treated with aqua regia at 70 (Cent.) 
 until the platinum is entirely dissolved. The aqua regia 
 must be renewed occasionally for 12 or 15 hours, or until 
 it is no longer coloured. It is best to perform this opera- 
 tion in a large beaker, and to place a cover over it to pre- 
 vent loss. The solution must be decanted with the greatest 
 care from the metallic spangles of the osm-iridium and the 
 sand which remain at the bottom of the beaker. If neces- 
 sary it may be filtered, but as little as possible of the 
 osmide must be allowed to go on the paper. The insoluble 
 residue must be washed by decantation, then dried and 
 weighed, after having added what remained on the filter- 
 By subtracting the weight of the residue from the weight 
 
DEVILLE AND DEBRAY's PROCESS. 797 
 
 of the sand obtained in the former operation, we obtain the 
 weight of the osm-iridium. For instance, in the California^ 
 ore we had : 
 
 Milligr. 
 
 Osm-iridium and sand 81 
 
 Sand .59 
 
 Osrn-iridium ,22 
 
 The button obtained in estimating the sand might be 
 employed in this operation. In that case it is necessary to 
 dissolve out the silver with nitric acid, and then proceed 
 with the residue, as we have just directed. 
 
 3. Platinum and Iridium. The solution in aqua regia 
 obtained in the last operation is evaporated to dryness at a 
 low temperature, and the residue is re-dissolved in a small 
 quantity of water (if it should not entirely dissolve in the 
 water, some more aqua regia must be added, and the evapo- 
 ration repeated), to which is added about twice as much 
 pure alcohol ; lastly, we add a great excess of sal-ammoniac 
 in crystals. The whole is now slightly warmed to com- 
 plete the solution of the sal-ammoniac ; it is then stirred, 
 and afterwards set aside for 24 hours. The orange-yellow, 
 or even reddish -brown, precipitate which is formed con- 
 tains the platinum and the iridium, but some remains in 
 the solution. The precipitate must be thrown on a filter 
 and washed with alcohol. Afterwards the filter is dried 
 in a platinum crucible, placed, for greater safety, within a 
 larger one, and then heated by degrees to low redness. 
 The crucibles are now uncovered, and the filter is burnt at 
 the lowest possible temperature. Once or twice after the 
 incineration of the filter a piece of paper saturated with 
 turpentine should be introduced into the crucible, by 
 which means the iridium oxide will be reduced, and the 
 expulsion of the last traces of osmium will be effected. 
 The crucible is now heated to whiteness until it no longer 
 loses weight, or the reduction is finished in a current of 
 hydrogen. 
 
 The liquid separated from the platinum -yellow by fil- 
 tration is evaporated until the ammonium chloride crys- 
 tallises in great quantity. It is allowed to cool, is then 
 
798 THE ASSAY OF PLATINUM. 
 
 decanted, and on a filter is collected a small quantity of 
 a deep violet-coloured salt, which is the iridium ammonio- 
 chloride mixed with a little of the platinum salt. This is 
 first washed with a solution of sal-ammoniac, and then with 
 alcohol. The salt is then ignited, and if necessary reduced 
 by hydrogen like the platinum salt. The mixture of pla- 
 tinum and iridium obtained by the two reductions is then 
 weighed. The two metals are now digested at about 40 
 or 50 (Cent.) in aqua regia, diluted with about 4 or 5 
 times its weight of water the aqua regia being renewed 
 until it is no longer coloured. The residue is pure iridium. 
 To obtain the weight of the platinum the weight of the 
 iridium is subtracted from that of the mixture of the two. 
 This method of separating the two metals is very accurate 
 If the aqua regia used be weak, and the contact with it 
 prolonged. 
 
 4. Palladium , Iron, and Copper. The liquid charged 
 with sal-ammoniac and alcohol, from which the platinum 
 .and iridium have been separated, is evaporated to get rid 
 of the alcohol, and then treated with an excess of nitric 
 acid, which transforms the ammonium chloride into nitro- 
 gen and hydrochloric acid. It is now evaporated almost 
 to dryness. The residue is removed to a covered porcelain 
 crucible, which is weighed with great care. When the 
 matter is dry it is moistened with concentrated ammonium 
 sulphide, and afterwards dusted over with 2 or 3 grammes 
 of pure sulphur. When dry this crucible is placed within 
 a larger one of clay, and surrounded with pieces of wood 
 charcoal. The two, covered, are now set in a cold furnace 
 which is filled up with charcoal, and the fire is lighted 
 at the top to avoid the projection of any matter from the 
 crucible if it were too quickly heated. After reaching a 
 bright red heat the crucibles are allowed to cool. The 
 porcelain crucible now contains palladium in a metallic 
 state, with the sulphides of iron and copper, and also the 
 gold and rhodium. This mixture is moistened with con- 
 centrated nitric acid, which, after prolonged digestion at 
 70, dissolves the palladium, iron, and copper, forming at 
 the same time a little sulphuric acid. The solution of the 
 
GUYABDS PEOCESS. 799 
 
 nitrate is poured off from the residue, which is washed by 
 decantation, and the solutions and washings are evaporated 
 to dryness, and then calcined at a strong red heat. In this 
 way the palladium is reduced, and the iron and copper 
 pass to the state of oxides, which are easily separated from 
 the palladium by means of strong hydrochloric acid. The 
 palladium remains in the crucible, in which it is again 
 strongly ignited and then weighed. 
 
 The iron and copper chlorides are now evaporated to 
 dryness at a temperature but little above 100 (Cent.), and 
 are then treated with ammonia. The ferric chloride, having 
 lost nearly all its acid, has become insoluble ; but the copper 
 chloride is readily dissolved, and may be filtered from the 
 iron, which is washed, ignited, and weighed. The copper 
 solution is now evaporated almost to dryness, and then mixed 
 with excess of nitric acid, and heated to drive off the am- 
 monium chloride. Afterwards the copper nitrate is ignited 
 and weighed. The weight of the copper is always so small 
 that the hygroscopic water the copper oxide may absorb 
 may be neglected. 
 
 5. Gold and Platinum. The residue insoluble in nitric 
 acid is weighed and treated with very dilute aqua regia 
 which takes up the gold, and sometimes, but very rarely, 
 traces of platinum. To ascertain if platinum be present, 
 evaporate to dryness, and re-dissolve by alcohol and am- 
 monium chloride. If any platinum-yellow remain, it must 
 be ignited and weighed. The difference in the weight of 
 the porcelain crucible before and after the treatment by 
 aqua regia gives the weight of the gold, from which, if 
 any be found, the weight of the platinum must be deducted. 
 
 6. Jthodium. The residue left in the crucible is rho- 
 dium, which must be reduced in a current of hydrogen. 
 
 We append the results of some analyses of platinum 
 ores, by MM. Deville and Debray. (See Table on next page.) 
 
 M. A. Guyard gives the following process for the ex- 
 traction of metals from platiniferous residues : 
 
 ' This process comprises three different operations, which 
 I will succinctly describe. 
 
 ' 1 . Solution of the Residues. The mother- liquors which 
 
800 
 
 THE ASSAY OP PLATINUM. 
 
 ANALYSES OF PLATINUM ORES FROM VARIOUS SOURCES. 
 
 
 Columbia 
 
 California 
 
 Oregon 
 
 Spain 
 
 Australia 
 
 Russia 
 
 Platinum 
 
 80-00 
 
 79-85 
 
 51-45 
 
 45-70 
 
 59-80 
 
 77-50 
 
 Iridium 
 
 1-55 
 
 4-20 
 
 0-40 
 
 0-95 
 
 2-20 
 
 1-45 
 
 Rhodium 
 
 2-50 
 
 0-65 
 
 0-65 
 
 2-65 
 
 1-50 
 
 2-80 
 
 Palladium 
 
 1-00 
 
 1-95 
 
 0-15 
 
 0-85 
 
 1-50 
 
 0-85 
 
 Gold . 
 
 1-50 
 
 0-55 
 
 0-85 
 
 3-15 
 
 2-40 
 
 (*) 
 
 Copper 
 Iron . 
 
 0-65 
 7-20 
 
 0-75 
 4-95 
 
 2-15 
 4-30 
 
 1-05 
 6-80 
 
 1-10 
 4-30 
 
 2-15 
 9-60 
 
 Osm-iridium 
 
 1-40 
 
 4-95 
 
 37-30 
 
 2-85 
 
 25-00 
 
 2-35 
 
 Sand . 
 
 4-35 
 
 2-60 
 
 3-00 
 
 35-95 
 
 1-20 
 
 1-00 
 
 Osmium and loss 
 
 
 
 0-05 
 
 
 
 0-05 
 
 0-80 
 
 2-30 
 
 
 100-15 
 
 100-00 
 
 100-25 
 
 100-00 
 
 100-00 
 
 100-00 
 
 remain after the precipitation of platinum by sal-ammoniac 
 come from solutions of crude or commercial platinum. 
 They always contain iron, mostly produced from the iron 
 sulphate used for the precipitation of gold, lead, copper, 
 palladium, platinum, and especially rhodium. These mother- 
 liquors are acidulated by hydrochloric acid, and are then 
 ready to be investigated. To recall their composition, I 
 shall distinguish them here only as residues in solution. It 
 need only be mentioned that iron, which is generally used 
 for the precipitation, must be avoided. 
 
 ' Solid residues are melted at once with three times 
 their weight of a mixture of equal parts of soda and sodium 
 nitrate. The fusion is effected at a bright red heat in a 
 thick iron vessel. It is accomplished without bubbling or 
 projection, and requires about an hour. During the last 
 twenty minutes the mass must be constantly stirred with 
 an iron spoon. The operation is extremely simple. 
 
 4 These residues contain osm-iridium, unattackable by 
 all chemical agents, attackable osmide, some grains of triple 
 alloy of platinum, iridiurn, and rhodium, which aqua regia 
 will not dissolve, but which nitre completely oxidises and 
 completely breaks up. They also contain the gangue 
 characteristic of platinum ores quartz, silicates of all 
 bases, titanates, hyacinth, &c., &c. 
 
 ' The mixture I make use of oxidises all that is oxidis- 
 
 * Gold, if any, counted in the loss. 
 
GUYAED S PEOCESS. 801 
 
 able, and breaks up the gangue, which it partly dissolves. 
 The melted mass contains all the bodies above mentioned, 
 besides a large quantity of iron oxide taken from the sides 
 of the vessel in which the operation is performed. The 
 fused mass is poured into cast-iron moulds. When solid 
 it is broken into fragments and boiled with sufficient water 
 to obtain a strong solution of soda, capable of holding all 
 the gelatinous acids in solution. It also contains osmium 
 in the state of osmiate.* It is filtered from insoluble 
 matter, and then supersaturated with hydrochloric acid. 
 The insoluble oxides are freed by washing from the excess 
 of alkali, and are then dissolved in aqua regia. 
 
 ' This solution contains iron, copper, lead, iridium, 
 rhodium, platinum, and ruthenium. It is separated from 
 the undissolved osmide, evaporated to expel the excess of 
 aqua regia and dissolved in water and hydrochloric acid. 
 
 4 2. Precipitation of Liquids by Sulphuretted Hydrogen. 
 Liquids obtained as above are ready for precipitation by 
 .hydrosulphuric acid. 
 
 4 The apparatus in which all the liquids are precipitated 
 is composed of a sulphuretted hydrogen gas generator by 
 the action of sulphuric acid on iron sulphide. This gene- 
 rator communicates with four or five large earthenware 
 jars holding about seventy litres, arranged precisely as in 
 Wolffs apparatus. A special tube conducts to each of 
 them the vapour destined to heat the liquid which they 
 contain. 
 
 c The whole apparatus is enclosed in a well-fitted wood 
 stove placed near a chimney, with which it communicates. 
 As to the small quantities of unabsorbed gas, they are 
 conducted into the chimney, where the fire creates a strong 
 draught. By this means, also, all smell is avoided during the 
 precipitation ; but after the operation air is forced through 
 the apparatus from large gasometers. It expels the hydro- 
 sulphuric acid which saturates the mother-liquors, and 
 these can then be manipulated free from smell, f 
 
 * This solution is separately precipitated by hydrosulphuric acid. Osmium 
 sulphide is thus isolated. 
 
 t A carbonic acid generator may be substituted for the gasometers and the 
 air with no difference in the result. 
 
 3F 
 
802 THE ASSAY OF PLATINUM. 
 
 ' The experiment is carried on during the precipitation 
 in the following manner : when the generator begins to 
 disengage gas, the temperature of the liquids is raised to 
 about 70. This temperature is maintained for nearly fif- 
 teen hours, that being the time required for the complete 
 precipitation of the sulphides, which collect better under 
 the influence of heat. The operation is concluded when 
 there remains but a very slight yellow tint in the mother- 
 liquor, arising from the presence of a little soluble iridium 
 sulphide. This mother-liquor is poured from the precipi- 
 tated sulphides into a vessel with pieces of iron, which 
 takes off a little of the iridium. The sulphides are filtered 
 through linen filters. 
 
 ' 3. Purification and Treatment of the Sulphides. The 
 mass of sulphides thus separated from the iron and from 
 all other bodies not precipitated by the sulphuretted gas, 
 contains, in addition to the sulphides of the platinum 
 metals, a large proportion of sulphur and the sulphides 
 of copper and lead. To get rid of these bodies, I have 
 thought of concentrated sulphuric acid, which changes 
 them to sulphurous acid and sulphates, while it does not 
 act on the sulphides of the precious metals. This refining 
 can be effected in an iron vessel ; but Mr. Matthey, who 
 neglects nothing to insure the certainty and exactness of 
 the results, makes use of platinum. 
 
 ' When, after prolonged boiling, no more sulphurous 
 acid is given off, the refining is complete. 
 
 ' The mass of sulphides, diluted with a quantity of 
 water, is thrown on filters, and thoroughly w r ashed, until 
 ammonia no longer shows any trace of copper or iron in 
 the filtered liquid. 
 
 ' At this point the precious metals are entirely freed 
 from iron, which is so detrimental to them, and from 
 copper, and contain only a little lead sulphate, which 
 separates by itself during an ulterior reaction. They are 
 then, moreover, in a condition to be dissolved by simple 
 nitric acid or by aqua regia, and this is not their least 
 valuable condition. 
 
 ' Treatment of the Sulphides. The sulphides are next 
 
ANALYSIS OF OSM-IRIDIUM. 803 
 
 dissolved in aqua regia, which should not be previously 
 prepared, because its action on sulphates is sudden and 
 energetic ; it heats so rapidly, and the disengagement of 
 gas is so great, that, were it previously prepared, it would 
 certainly be thrown from the vessels. 
 
 ' I add then moderately strong cold nitric acid, and 
 add it gradually, because its action is strong. A quantity 
 of rutilant vapours are disengaged. Hydrochloric acid is 
 added when the effervescence ceases. It is then gradually 
 heated to boiling, which is necessary to obtain a complete 
 solution. 
 
 ' The solution is poured from the deposited lead 
 chloride, and the ordinary method with sal-ammoniac is 
 used to separate the different metals it contains. Experi- 
 ments on large quantities of material have fully proved 
 the advantages of this process.' 
 
 Analysis of Osm-iridium. Wohler's method of resolv- 
 ing osm-iridium consists in passing moist chlorine over 
 the ore mixed with common salt and heated to low red- 
 ness in a glass or porcelain tube. This method is invalu- 
 able in analysis, and gives excellent results in working the 
 ore upon a small scale. In all cases, however, several 
 repetitions of the process are necessary for complete 
 resolution or reduction to a soluble form. On the other 
 hand, it can scarcely be doubted that this method could be 
 advantageously employed upon the large scale, if vessels 
 of porcelain of large size and of a proper shape could 
 be obtained. Such vessels might be constructed in the 
 form of long and flattened ellipsoids, furnished at each 
 extremity with wide tubes several inches in length, and 
 would be of great utility in various chemical processes. 
 No process of fusion with oxidising agents compares with 
 Wohler's method in point of elegance, as no iron or other 
 impurities afterwards to be removed are introduced by 
 the process itself. 
 
 Claus's method of resolving the ore consists in fusing 
 for an hour, at a red heat, a mixture of one part of ore 
 with one part of caustic potash and two of saltpetre. The 
 fused mass is to be poured out upon a stone, allowed to 
 
 3 F 2 
 
804 THE ASSAY OF PLATINUM. 
 
 cool, broken into small pieces or powdered, and then 
 introduced into a flask, which is to be filled with cold 
 water and allowed to stand for twenty-four hours. The 
 clear deep orange-red solution of potassium osmiate and 
 rutheniate is then to be drawn off by means of a syphon, 
 and the black mass remaining again washed in the same 
 manner. The finely divided oxidised portion of the in- 
 soluble matter may now be separated from, the unattacked 
 ore by diffusion in water and pouring off, after the sub- 
 sidence of the heavier ore. The unattacked ore is then 
 to be fused a second time with potash and saltpetre and 
 treated as before. Clans asserts that he has been able in 
 this manner to resolve the Siberian osm-iridium completely 
 in two operations. 
 
 Dr. Wolcott Gibbs, to whom the chemistry of the 
 platinum metals is so greatly indebted, recommends the 
 following process for the analysis of osm-iridium : The 
 ore, which is usually very impure, is in the first place- 
 to be fused with three times its weight of dry sodium 
 carbonate. The fused mass after cooling is to be treated 
 with hot water, to remove all the soluble portions, and 
 then the lighter portions are to be separated by washing 
 from the heavy unattacked ore. In this manner the 
 greater part of the silica and other impurities present rnay 
 be removed. A previous purification of this kind is not 
 indispensable, and may be omitted altogether when the 
 ore is in plates or large grains ; but it is very desirable 
 when the ore is in fine powder, and greatly facilitates the 
 subsequent action of the oxidising mixture. By cutting 
 off- the top of a mercury bottle a wrought-iron crucible is 
 obtained, in which 600 grms. of osm-iridium may be fused 
 at one operation with potash and saltpetre as above. 
 There is usually little or no foaming, and if any occur it 
 may easily be checked by stirring with an iron rod. No 
 sensible quantity of osmic acid is given off during the 
 process, which with a little care is entirely free from 
 danger. In this manner 1,500 grms. of ore have been 
 worked up in a few hours in three successive operations. 
 The fused mass is to be broken into pieces with a hammer, 
 
 
WOLCOTT GIBBS'S PROCESS. 805 
 
 and placed in a clean iron pot. Boiling water, contain- 
 ing about one-tenth of its volume of strong alcohol, is 
 then to be added, and the whole is to be boiled over an 
 open fire until the fused mass is completely disintegrated. 
 The potassium osmiate is, in this manner, reduced to 
 osmite, while the potassium rutheniate is completely 
 decomposed, the ruthenium being precipitated as a black 
 powder. It is advantageous, after boiling for some time, 
 to pour off the supernatant liquid with the lighter por- 
 tions of the oxides, and boil a second time with a fresh 
 mixture of alcohol and water. In this manner we obtain 
 a solution of potassium osmite, a large quantity of black 
 oxides, and a heavy black and coarse powder. This last 
 consists chiefly of undecomposed ore, mixed with a small 
 quantity of the iridium oxides, &c., with scales of iron 
 oxide from the crucible, and, if the ore has not been 
 previously purified, with the impurities of the ore itself. 
 The greater specific gravity of this residual mass renders 
 it very easy to pour off from it the mixture of black 
 oxides with the solution of osmite of potash and alkaline 
 salts. . This solution with the suspended powder is to be 
 poured into a beaker and allowed to settle. The heavy 
 black powder remaining in the iron pot is then to be 
 perfectly dried over the fire, and fused a second time with 
 potash and saltpetre as before. The fused mass is to be 
 treated exactly as after the first fusion. The heavy por- 
 tions remaining after this operation may be fused a third 
 time with the oxidising mixture. When, however, the ore 
 has been previously purified by fusion with sodium carbon- 
 ate, or when it was originally in the form of clean scales, 
 the heavy portion remaining after two successive oxidations 
 will be found to consist chiefly of scales of iron oxide. 
 
 The solutions containing potassium osmite and alka- 
 line salts are to be carefully drawn off by a syphon from 
 the black oxides which have settled to the bottom of the 
 Containing vessels. The oxides may then be washed with 
 hot water containing a little alcohol, and introduced into 
 a capacious retort. By this process, when carefully 
 executed, no trace of osmic acid escapes an advantage 
 
800 THE ASSAY OF PLATINUM. 
 
 not to be despised, as the deleterious effects of this body 
 upon the lungs have not been exaggerated, and too much 
 care cannot be taken to avoid inhaling it. 
 
 The solution of alkaline salts contains only a portion 
 of the osmium in the ore. The other portion exists in the 
 mixture of oxide, and must be separated by distillation. 
 For this purpose the retort should be provided with a 
 safety-tube, passing through the tubulure, and with a 
 receiver kept cold, and connected by a wide bent tube 
 with a series of two or three two-necked bottles contain- 
 ing a strong solution of caustic potash with a little alcohol, 
 and also kept cold. All the tubulures and connections 
 must be made perfectly tight. Strong hydrochloric acid 
 is then to be cautiously poured into the retort, through 
 the safety-tube, in small portions at a time. The reaction 
 which ensues is often violent ; great heat is evolved, and 
 a portion of the osmic acid distils over immediately, and 
 condenses in the receiver in the form of colourless needles. 
 When a large excess of acid has been added, the action 
 entirely ceased, and the retort become cold, heat may 
 be applied by means of a sand-bath. The osmic acid 
 gradually distils over, and condenses in the receiver and 
 in the two-necked bottles. Especial care must be taken 
 that the beak of the retort is not too small at the ex- 
 tremity, as it may otherwise become completely stopped 
 up with the condensed osmic acid. The same applies to 
 the tubes which connect the receivers and two-necked 
 bottles. The distillation should be continued for some 
 time after osmic acid ceases to appear in the neck of the 
 retort ; when this has once become hot, the acid condenses, 
 and passes into the receiver in the form of oily drops. 
 
 When the distillation is finished, the retort is to be 
 allowed to cool, and then separated from the receiver, 
 which is to be immediately closed with a cork. By gently 
 heating the receiver in a water-bath, the contained osmic 
 acid may be driven over into the two-necked bottles, 
 where it condenses in the alkaline solution, and is reduced 
 by the alcohol to potassium osmite. The solution thus 
 obtained may be added to that obtained directly from the 
 
NELSON w. PERRY'S PROCESS. 807 
 
 fused mass of ore, and, on evaporation in a water-bath and 
 cooling, will yield crystals of potassium osmite, the salt 
 being but slightly soluble in strong saline solutions. The 
 mother- liquor from the crystals contains only traces of 
 osmium, and may be thrown away as worthless. 
 
 The dissolved portions drawn off from the retort have 
 a very dark brown-red colour. The solution is to be 
 evaporated to dryness, re-dissolved in hot water and again 
 evaporated, after adding a little hydrochloric acid, and 
 this process repeated till no smell of osmic acid can be 
 perceived. A cold and saturated solution of potassium 
 chloride is then to be added in large excess. This dissolves 
 the iron and palladium chlorides which may be present, 
 leaving platinum, iridium, rhodium, and ruthenium as 
 double chlorides, insoluble in a strong solution of the 
 alkaline chloride. 
 
 The undissolved mass is to be well washed with a 
 saturated solution of potassium chloride, which is prefer- 
 able to sal-ammoniac. In this manner nearly the whole 
 of the iron and palladium may be removed, while any 
 insoluble impurities contained in the ore remain with the 
 mixed double chlorides. 
 
 For the separation of osmium from the other metals of 
 the group, the best plan seems to be the one which is 
 universally employed namely, the volatilisation of the 
 osmium, in the form of osmic acid. 
 
 Mr. Nelson W. Perry (' Engineering and Mining Jour- 
 nal ') proposes the following method for the assay of 
 platinum alloys containing base metal, silver, platinum, 
 gold, and osm-iridium. 
 
 Charge, platinum alloy 200 milligrammes, pure silver 
 150 milligrammes, or sufficient to produce perfect cu- 
 pellation. 
 
 Wrap charge in sheet lead and cupel. Weigh button. 
 Loss = base metal. 
 
 Flatten button, anneal, roll out thin, anneal again, and 
 make into cornet as in gold bullion assay. Introduce 
 cornet into parting flask and part with concentrated 
 sulphuric acid. Wash, anneal, and weigh. Loss from 
 
808 THE ASSAY OF PLATINUM. 
 
 previous weight = silver in original alloy + silver added 
 for cupellation. 
 
 Alloy cornet with at least twelve times the amount of 
 silver that there is platinum present, and as before, form 
 cornet, and part first with nitric acid, sp. gr. 1*16, and 
 then nitric acid, sp. gr. 1-26. Wash thoroughly, anneal in 
 annealing cup, and weigh. 
 
 Treat residue with aqua regia, obtain gold by loss, the 
 residue is osm-iridium. 
 
 Time to complete assay in duplicate, 2 hours 45 min. 
 
 The quality of the silver added should at least be 
 sufficient, so that after the addition of the silver the alloy 
 will be to the gold as 3 : 1. 
 
 As platinum and osm-iridium add greatly to the in 
 fusibility of the compound, silver in sufficient quantity 
 must be added to prevent ' freezing ' and give a perfect 
 cupellation. Any large excess over these requirements is 
 to be avoided, first, because the residue, after parting, will 
 in that case be non-adherent, and in a more or less fine 
 state of subdivision, which .may occasion loss in washing 
 by decantation ; second, the larger the button cupelled, 
 the more difficult it is to obtain a good cupellation, and 
 the greater loss of silver during the process. It may, 
 for this reason, sometimes be necessary to use only 100 
 milligrammes of the alloy for assay instead of 200 milli- 
 grammes as above. 
 
 The cupellation should take place at a moderate 
 temperature, until near the ' blick,' when the assay should 
 be thrust back into the hottest part of the furnace to 
 prevent ' freezing.' The button must remain in the muffle 
 until all the lead is gone. 
 
 In parting with sulphuric acid boil for several minutes. 
 In other respects this operation is identical with -the gold 
 bullion assay. Any large excess of silver over twelve 
 times the amount of platinum in alloy is to be avoided, as 
 it causes the residue, after parting, to be too fine and float, 
 thereby occasioning loss in washing. Insufficient silver is 
 even worse, as the platinum will then be only incompletely 
 dissolved. 
 
CHAPTEK XIX. 
 
 THE ASSAY OF BISMUTH. 
 
 THE following varieties of bismuth ores are met with, but 
 are somewhat rare : 
 
 Native Bismuth. 
 
 Bismuth Oxide. 
 
 Bismuth Sulphide. 
 
 Bismuth Persulphide. 
 
 Cupriferous Bismuth Sulphide. 
 
 Plumbo-cupriferous Bismuth Sulphide. 
 
 Plumbo-argentiferous Bismuth Sulphide. 
 
 Native Bismuth possesses a tolerably bright metallic 
 lustre ; its colour yellowish-white, often iridescent. It 
 fuses in the candle flame. It is generally found in small 
 amorphous lamellar masses, yet it occasionally occurs in 
 acute rhomboidal as well as cubical and octahedral 
 crystals. 
 
 This substance does not seem to form veins by itself, 
 but generally accompanies other minerals, particularly 
 those of cobalt, nickel, arsenic, and lead. 
 
 To within a recent period the chief source of the com- 
 mercial product has been native bismuth. But this limited 
 source is becoming well-nigh exhausted, whilst the de- 
 mand for this metal, especially in a great state of purity, 
 is increasing every day. It has thus become a matter of 
 necessity to look for fresh fields of exploration, for new 
 deposits : and as bismuth ores of every description mixed 
 up with other ores of various kinds are now used for the 
 extraction of bismuth, the assay of this metal has lost some 
 of its former simplicity. Mr. Hugo Tamm has done more 
 than any metallurgist towards perfecting the assay of 
 
810 THE ASSAY OF BISMUTH. 
 
 bismuth, and from his papers on this subject in the 
 4 Chemical News,' Nos. 639 and 640, the following method 
 of assay is condensed : 
 
 Assaying of Bismuth Ores. Whenever the ore to be 
 tried is of a simple nature, is free from admixture with 
 other ores, and contains bismuth in the metallic state, or 
 in the state of sulphide, of oxide, or of carbonate, or, as 
 sometimes occurs, consists of a mixture of oxide, carbon- 
 ate, sulphate, and oxychloride, the assaying of bismuth is 
 reduced to the very simple operation of mixing the ore with 
 as fusible a flux as can easily be obtained, to which a 
 reducing substance, generally charcoal powder, is added 
 in proper quantity. 
 
 It is of course useless to lay down particular rules 
 concerning the nature or the quantity of the flux, and of 
 the reducing substance to be employed in this operation ; 
 indeed, it is not advisable to do so, and it is by far the 
 best to be guided by the nature of the materials at hand, 
 and by the results of a few trials with varied proportions 
 of flux and of the reducing agent ; the aim of the assayer 
 being the highest amount of metal that can be obtained in 
 a given instance. Still, one of the best fluxes, as well as 
 one of the most simple, consists of a mixture of two parts 
 of potassium or sodium carbonate, and one part of sodium 
 chloride, to which a proper amount of red argol or of 
 potassium cyanide on the small scale, and powdered 
 charcoal on the large scale, are added. 
 
 Assaying of Bismuth in Ores containing a large amount 
 of Copper. The problem of the direct separation of bis- 
 muth from ores containing large proportions of copper has 
 hitherto been one of difficulty, and its solution, which was 
 of considerable importance, offered great interest. The 
 difficulty consisted chiefly in the fact that both copper 
 and bismuth behave, in nearly every instance, in an 
 identical manner with docimastic reagents ; but Mr. 
 Tamm has happily hit upon a most simple and practical 
 means of effecting the direct separation of those two 
 metals. 
 
 The chief kinds of ores containing both bismuth and 
 
ASSAYING BISMUTH ORES WITH MUCH COFFEE. 811 
 
 copper are the bismuth-copper pyrites or sulphuretted 
 ores, and the double bismuth and copper oxides or car- 
 bonates, or oxidised ores. 
 
 Both kinds of ores may be, and generally are, con- 
 taminated with other metals, but these foreign metals con- 
 stitute only, as a rule, a small fraction of the w r hole, and 
 the problem of their elimination will be given further on.. 
 
 The reaction upon which the separation of bismuth 
 from copper is founded consists in the fact that, in pre- 
 sence of alkaline fluxes, carbonaceous reagents, and, of 
 course, among them carbon itself, reduce bismuth sul- 
 phide to the metallic state, while copper sulphide is not 
 reduced. 
 
 In the treatment of sulphuretted ores, both metals 
 being already in the state of sulphides, all that is required 
 is to run them down with a mixture of potassium carbon- 
 ate or soda and salt, to which a little flowers of sulphur 
 or ground sulphur, and charcoal or any other carbonaceous 
 substance, are added. 
 
 In this operation metallic bismuth is extracted quite 
 easily, and the metal thus obtained is tolerably free from 
 copper. It is recommended to add a little sulphur in 
 order to insure a complete sulphurisation of copper during 
 the whole of the operation, and especially to prevent any 
 desulphurisation of copper by the alkali, and, consequently,, 
 to prevent, as much as possible, this metal from being 
 reduced. 
 
 With oxidised ores the operation is very similar in 
 every respect to the one just described, and it differs from 
 it only in the amount of sulphur used, which is greater 
 in this instance, since the whole of the metals have to be 
 sulphurised. 
 
 Three parts of the ore are mixed with from two to 
 three parts of a flux composed of 
 
 Sodium carbonate .... 5 parts 
 
 Salt 2 
 
 Sulphur 2 
 
 Charcoal powder . . . . .1 part 
 
 Both the composition of the flux and the amount to be 
 used may be altered with advantage to suit each particular 
 
812 THE ASSAY OF BISMUTH. 
 
 case. A few synthetical trials, in the hands of a person 
 accustomed to metallurgical operations, are all that are 
 required to make the best use of this reaction. 
 
 In general, it is to be observed that the amount of flux 
 and of reagents required for the assaying may be con- 
 siderably reduced when the operation is carried on on a 
 larger scale. On the other hand, it is scarcely worth 
 while mentioning that, in the operation of assaying, potas- 
 sium cyanide forms an admirable substitute for carbon. 
 
 During the process of extracting bismuth by means of 
 sulphur and carbon there is a loss of about 8 per cent, of 
 the bismuth contained in the ore. This loss is unavoidable, 
 but there is a more than proportionate loss of the metals 
 arsenic, antimony, and lead, which in this operation are 
 reduced with bismuth, and the crude metal obtained by 
 this process is not so impure as the corresponding metal 
 obtained by the direct reduction of the oxidised ores ; 
 besides the whole of the copper remains in the slag. 
 
 Whenever the sulphur-carbon process is employed, the 
 use of iron stirrers must be carefully avoided, for the reason 
 that copper sulphide is rapidly reduced to the metallic state 
 by this metal, especially in the presence of alkalies. 
 
 This process for the separation of bismuth from copper 
 will be found chiefly useful and important for the separa- 
 tion of bismuth in minerals containing large quantities of 
 copper. When, on the contrary, this metal exists only in 
 smaller proportions, it is more advantageous to run down 
 the whole of the metals, and to separate them afterwards 
 in the special operations of refining. But it is recom- 
 mended that the sulphur-carbon process be used for the 
 treatment of the somewhat abundant ores of bismuth 
 formed of bismuth and lead oxides, and small proportions 
 ofarsenious acid and antimonious acid, with a lit tie copper 
 oxide ; for there is as yet no direct means of smelting pure 
 bismuth from ores containing large proportions "of lead, 
 but it has been observed that bismuth' extracted by the 
 sulphur process contains less lead than the corresponding 
 metal obtained direct from the oxidised ore. The same 
 remark applies to arsenic and antimony, and this is in 
 
PURIFICATION OF BISMUTH. 813 
 
 accordance with the behaviour of the sulphides of these 
 metals with alkaline sulphides. 
 
 Refining Crude Bismuth. The various ores of bismuth 
 above described, whether sulphuretted or oxidised, are 
 seldom formed of bismuth and iron only, or of only 
 bismuth, copper, and iron. They nearly always are con- 
 taminated by various proportions of lead, arsenic, or anti- 
 mony, metals which are reduced with bismuth, partially at 
 least, whatever process has been used for the extraction of 
 bismuth, and besides, the metal obtained by the sulphur 
 process from copper bismuth ores still contains a small 
 quantity of copper, which it is important to remove. 
 
 Bismuth extracted by any process is so generally free 
 from iron that no notice need be taken of this metal, which 
 remains wholly in the slags. 
 
 The fracture of good bismuth and that of its various 
 alloys is so characteristic that it is not often necessary to 
 have recourse to tests in order to determine what particular 
 processes will have to be used for the refining of the crude 
 metal. 
 
 Pure bismuth is tougher than most of its alloys. Its 
 fracture is bright, and it possesses a fine reddish colour. 
 Bismuth containing arsenic gives a beautiful fracture con- 
 sisting of large laminae of a whiter colour than that of pure 
 bismuth. Copper mixes with bismuth without alloying 
 with it, and is almost always discernible. The fracture of 
 bismuth containing antimony is dull and is mostly composed 
 of very small crystals. Lead does not prevent bismuth from 
 crystallising in large crystals, but these crystals are studded 
 all over with fine crystals. Sulphur imparts a black tinge 
 to metallic bismuth. 
 
 To these appearances, which almost suffice to an ex- 
 perienced eye, may be added a few simple tests. 
 
 It is difficult to detect arsenic in the presence of a large 
 quantity of bismuth by means of reagents, and the most 
 simple way of detecting this substance is to heat the bis- 
 muth on charcoal, with the oxidising flame of the blow- 
 pipe. Very small quantities of arsenic may be detected 
 by the garlic odour evolved. 
 
814 THE ASSAY OF BISMUTH. 
 
 To detect copper the metal is dissolved in nitric acid, 
 the solution is supersaturated by ammonia, and filtered. 
 The blue colour of the filtrate indicates the presence of 
 copper. 
 
 When bismuth dissolves in strong nitric acid, with the 
 formation of a cloudy white precipitate which does not 
 disappear on the addition of w r ater, it is because antimony 
 is present. 
 
 When bismuth dissolves in strong nitric acid, with the 
 formation of a very white granular or crystalline precipi- 
 tate which dissolves freely on the addition of water, the 
 presence of lead is indicated. 
 
 But to detect with absolute certainty the presence of 
 even very small proportions of lead, the metal is dissolved 
 in nitric acid. The solution is supersaturated with am- 
 monia, and re-acidulated with the smallest amount of hydro- 
 chloric acid which will give a clear liquor. This liquor is 
 then precipitated by a large excess of boiling water. Water 
 must be added until no further precipitation takes place. 
 The whole is then filtered, and the filtrate is saturated with 
 a mixture of ammonia and ammonium carbonate ; when 
 a yellowish-white precipitate is formed it is because lead 
 exists in the bismuth. 
 
 It may be useful to submit the metal to be refined to 
 these various tests in order to ascertain beforehand which 
 refining process should be used. But it is essential to apply 
 each test to the refined metal, so as to verify its degree of 
 purity. 
 
 PURIFICATION OF THE KEDUCED BISMUTH. 
 
 Purification of Bismuth from Arsenic. The separation 
 of bismuth from arsenic is founded on the almost absolute 
 want of affinity of bismuth for iron, on the readiness with 
 which arsenic combines with iron, and on the fact that the 
 iron arsenide thus formed does not alloy with bismuth. 
 This operation is conducted in the following manner : 
 
 Bismuth is melted at a relatively high temperature, at 
 a bright red heat, under cover of borax or fiux, to avoid 
 loss of bismuth by volatilisation, and strips of iron are 
 
PURIFICATION OF BISMUTH. 815 
 
 plunged into the molten metal. Iron is, according to the 
 technical expression, rapidly ' eaten away,' forming iron 
 arsenide, which rises to the surface of the metal. 
 
 When it is ascertained that fresh pieces of iron are no 
 longer attacked, the whole is allowed to cool. The iron 
 arsenide sets rapidly, and the bismuth, which is still fluid, 
 is poured out of the crucible into moulds. Singularly 
 enough, this process, which succeeds in perfection for the 
 separation of arsenic, is valueless when applied to the 
 separation of bismuth from antimony ; although, be it 
 noticed, the affinity of this metal for iron is very great. 
 Some antimony is removed by this process, but part of it 
 only, and it must be admitted that bismuth has as much, 
 or more, affinity for antimony than iron. 
 
 Purification of Bismuth from Antimony. The best way 
 of separating the two metals is to melt the alloy with a 
 quantity of bismuth oxide, equal to two and a half or three 
 times the weight of the antimony contained in the alloy. 
 The bismuth oxide is instantaneously reduced to the 
 metallic state, and antimony is liberated under the form of 
 antimony oxide, which combines with a little bismuth oxide, 
 and floats on the surface of the pure metal, whence it can 
 easily be removed. 
 
 This operation must be performed in clay crucibles, and 
 both carbon and iron must be carefully excluded, to avoid 
 any reduction of antimony oxide. The least traces of 
 antimony may be removed by this process without any 
 difficulty whatever. 
 
 Purification of Bismuth from Copper. When bismuth 
 ores contain only a small percentage of copper, and when 
 the ores are oxidised ores, it is advantageous to reduce 
 them at once by carbon and fluxes, without going through 
 the sulphurising process ; and, as a matter of course, all 
 the copper is alloyed with the bismuth. 
 
 On the other hand, bismuth extracted from copper 
 ores by the sulphur process contains, even in the best con- 
 ducted operation, a certain proportion of copper which 
 must be removed. This elimination has hitherto presented 
 very great difficulties, and could not be effected without 
 
816 THE ASSAY OF BISMUTH. 
 
 losing a large amount of bismuth. Mr. Tamm has, how- 
 ever, devised the following method, by melting the alloy 
 with potassium sulphocyanide. 
 
 The sulphocyanide is prepared during the process of 
 refining, by mixing together eight parts of potassium cya- 
 nide and three parts of flowers of sulphur. One part of 
 this mixture is thrown on to sixteen parts of the metal 
 melted at a low temperature. 
 
 A reaction soon takes place, by which the mass of the 
 metal is brought to a bright red heat, and, at the same 
 time, the sulphocyanide begins to burn vividly, throwing, 
 in every direction, showers of scintillating sparks emitting 
 a blue light. 
 
 The crucible is covered over, and great care must be 
 taken to prevent the heat from rising above the burning 
 point of the sulphocyanide, a temperature at which bismuth 
 sulphide begins to volatilise. 
 
 The reaction is allowed to exhaust itself, and, when 
 all is quiet, and after the metal has been well stirred with 
 a clay stirrer (iron must be avoided), the flux is allowed to 
 set, and the metal, which is still fluid, is poured out into 
 moulds. 
 
 Purification of Bismuth from Sulphur. The metal ob- 
 tained in the above operation contains some sulphur. To 
 remove this substance, the metal is melted with iron or 
 carbon ; the separation is thus effected easily. 
 
 The several processes here proposed are chiefly useful 
 for the refining of bismuth alloyed with one metal. 
 
 There is no dry method of refining by one process 
 bismuth alloyed with several metals ; but the succes- 
 sive use of these different methods can safely be recom- 
 mended. 
 
 Copper should be removed first, for the reason that 
 some lead, antimony, and arsenic are eliminated at the same 
 time. 
 
 Bismuth should next be freed from antimony, and, 
 lastly, from arsenic and sulphur. 
 
 Herr Thtirach (' Journal fur Prakt. Chemie,' P. S. 14, 
 315) precipitates with sulphuretted hydrogen from a hot 
 
VOLUMETRIC ASSAY OF BISMUTH. 817 
 
 solution, washes with hot water, heats for a considerable 
 time to 200-300 in a covered crucible, roasts in the open 
 crucible, and finally ignites strongly arid weighs as bismuth 
 oxide. 
 
 VOLUMETEIC ASSAY OF BISMUTH. 
 
 Mr. E. W. Pearson has given the following process for 
 the Yolumetric Assay of Bismuth. 
 
 Preparation of Standard Solution. '7135 grain of 
 pure crystallised potassium bichromate is dissolved in 
 100 grains of water. Call this solution the bichrome test 
 A. In a similar way, prepare a second solution, one-tenth 
 the strength of bichrome test A ; -07135 grain of potassium 
 bichromate, dissolved in 100 grains of water, will furnish 
 such a solution ; call it the bichrome test B. Bichrome 
 test C, one-tenth the strength of solution B, is also prepared 
 by dissolving -007135 grain of the potassium bichromate in 
 100 grains of water. 
 
 These figures can be multiplied to any convenient 
 number. These solutions will contain potassium bichro- 
 mate, in 100 grains of bichrome test A, equal to 1 grain of 
 bismuth ; in 100 grains of bichrome test B, equal to O'JL 
 grain of bismuth; and in 100 grains of bichrome test C, 
 equal to -01 grain of bismuth. 
 
 The bismuth should be in the form of nitrate, and the 
 solution kept hot during the experiment, as the precipi- 
 tated chromate collects more readily then; after com- 
 plete precipitation of the bismuth the solution will exhibit 
 a characteristic colour, produced by excess of potassium 
 bichromate. 
 
 By employing a standard solution of bismuth it has 
 been ascertained that 71*35 parts of potassium bichromate 
 are required to combine with 100 parts of bismuth. 
 
 For alloys of tin, lead, and bismuth (fusible metal) the 
 alloy, finely divided, is oxidised in a roomy flask with 
 nitric acid, the liquid is somewhat diluted, supersaturated 
 with ammonia, and digested for a long time and with -fre- 
 quent agitation with ammonium hydrosulphide, to which 
 
 3 G 
 
818 . THE ASSAY OF BISMUTH. 
 
 a little sulphur has been added. The tin is thus dissolved 
 as sulphide. The lead and bismuth sulphides are filtered 
 off and washed with cold water. The liquid containing 
 the tin sulphide is slightly supersaturated with dilute sul- 
 phuric acid and very gently warmed, the vessel being 
 loosely covered with paper till the odour of sulphuretted 
 hydrogen has disappeared. The yellow tin sulphide is 
 washed on a filter with cold water, and treated as directed 
 for the analysis of tin ore. 
 
 The lead and bismuth sulphides are dried, detached from 
 the filter, which is incinerated with the usual precautions, 
 the whole digested with nitric acid, and the solution con- 
 centrated at last with the addition of hydrochloric acid till 
 it is reduced to a small bulk, and the greatest part of the 
 lead is separated as chloride. The proportion of acid must 
 be so large that the liquid is not rendered turbid by a 
 slight addition of water. Sulphuric acid is added, and the 
 whole let stand, but with frequent stirring. After the lead 
 chloride has been converted into sulphate a little alcohol 
 of O80 sp. gr. is added, and the whole brought upon 
 a weighed filter, where it is washed, first with alcohol 
 acidulated with hydrochloric acid, and afterwards with 
 water. 
 
 In the filtrate all the bismuth is precipitated as basic 
 chloride by the addition of water in large excess, and fil- 
 tered off (the filtrate being tested with sulphuretted hydro- 
 gen), washed with cold water, dried, and melted at a 
 moderate heat with four parts potassium cyanide in a 
 covered porcelain crucible (Eammelsberg). 
 
 Mr. M. Patteson Muir (< Chem. News,' April 27, 1877) 
 finds the following solution a most sensitive test for bis- 
 muth : 12 grms. crystalline tartaric acid, and 4 grms. 
 stannous chloride are dissolved in caustic potash, producing 
 a clear liquid of a decidedly alkaline reaction, which 
 should remain clear at 60 to 70 C. To the liquid to be 
 tested is added a considerable quantity of tartaric acid. 
 It is warmed and made alkaline with caustic potash. A 
 few c.c. of the stannous chloride solution (called from its 
 
VOLUMETRIC ASSAY OF BISMUTH. 819 
 
 first discoverer Schneider's reagent) are added, and the 
 liquid warmed to 60 to 70 C. for a few minutes. If bis- 
 muth is present to the extent of one part in 210,000 parts, 
 a brownish-black colour is produced. Mercury must be 
 absent ; copper and manganese interfere slightly. 
 
 3 G 2 
 
820 
 
 CHAPTER XX. 
 
 THE ASSAY OF CHROMIUM. 
 
 THE principal ore of this metal which occurs in commerce 
 is known as chrome iron, or chrome iron ore. It is found 
 in amorphous masses of a brownish-black colour, approach- 
 ing an iron grey. Its fracture is uneven, sometimes la- 
 mellar ; and its powder is greyish. 
 
 The two following analyses will give a general idea of 
 its composition : 
 
 Chromium oxide .... 36'0 43'7 
 
 Ferric oxide 37'0 34-7 
 
 Alumina . . . . . 21-5 20-3 
 
 Silica . . . .* ... 5-0 2'0 
 
 99-5 100-7 
 
 Assay of Chrome Iron Ore. 
 
 Chrome iron ore, like native tin oxide, is very difficultly 
 decomposable by ordinary reagents. 
 
 Dr. Genth, of Philadelphia, who has had much experi- 
 ence in the analysis of chrome-iron ore, gives the following 
 process ; it is very trustworthy, although long and some- 
 what tedious. 
 
 Of the chrome ore, reduced to an impalpable powder, 
 put 0-5 gramme in a platinum crucible about 2 inches 
 high, nearly If inch wide, and holding 52 grammes of 
 water, and place upon it 6 grammes of pure fused potas- 
 sium bisulphate, and heat with the greatest care for about 
 15 minutes, at a temperature scarcely above the fusing 
 point of the bisulphate ; then the heat is gradually raised, 
 but not higher than to make the bottom of the crucible 
 red hot, and kept at this temperature from 15 to 20 
 minutes. Never permit the mass to rise to half the height 
 
ASSAY OF CHROME IRON ORE. 521 
 
 of the crucible. (If the fusion with potassium bisulphate 
 is done too rapidly a portion of the analysis is very apt to 
 be lost by spluttering, from the escape of sulphurous acid, 
 resulting from the oxidation of the ferrous oxide by the 
 sulphuric acid.) The mass begins now to fuse quietly, 
 and vapours of sulphuric acid go off more freely ; it should 
 then be kept at a red heat for about twenty minutes, and 
 the heat next raised as high as necessary to drive off the 
 second equivalent of sulphuric acid, and even to decom- 
 pose a portion of the iron and chromium sulphates. To 
 the fused mass add about three grammes of pure sodium 
 carbonate, and fuse the mixture, and then, by degrees, 
 keeping the temperature for about one hour at a dull red 
 heat, about the same quantity of saltpetre ; next heat for 
 fifteen minutes at a bright red heat. The fused mass is 
 dissolved in boiling water, filtered whilst boiling, and 
 washed with boiling water. 
 
 The insoluble residue, containing the greater portion of 
 the silicic acid, titanic acid, and alumina, the ferric oxide, 
 zirconia, and if the fusion has been conducted at a tem- 
 perature sufficiently high to convert the saltpetre into 
 caustic potash, and the above precautions have been used 
 ! all the magnesia, is re-dissolved in dilute warm hydro- 
 chloric acid, which generally dissolves it readily and com- 
 pletely, and rarely leaves un decomposed ore behind ; but 
 if so, this residue must invariably be fused in a small 
 Crucible as before, adding, after the separation of the in- 
 soluble portion, the solution containing the small quantity 
 of chromic acid to the first filtrate.* (The certainly less 
 troublesome method, to deduct the insoluble portion from 
 the original weight, is bad ; such residues have never the 
 .composition of the original ore.) The filtrate contains the 
 whole quantity of the chromium as chromic acid, some- 
 times a trace of manganic acid, small quantities of silicic 
 acid, alumina, and rarely titanic acid. To this solution 
 add an excess of ammonium nitrate, and evaporate over a 
 water-bath nearly to dryness, and until all the liberated 
 ammonia has been expelled. The precipitate, remaining 
 ron addition of water, contains the silicic acid, titanic acid, 
 
822 THE ASSAY OF CHEOMIUM. 
 
 alumina, and manganic oxide, which had gone into solution 
 with the chromic acid ; it is filtered off, and the filtrate 
 made strongly acid with sulphurous acid, carefully heated 
 to boiling, precipitated with a slight excess of ammonia, 
 boiled for a few minutes, and filtered. Dr. Genth says he 
 formerly acidulated the chromic acid solution by hydro- 
 chloric acid and then added sulphurous acid, but he several 
 times observed that, although an excess of sulphurous 
 acid had been used, a small portion of the chromic acid 
 escaped reduction, the filtrate from the ammonia pre- 
 cipitate being yellow. He has in vain tried to find the 
 reason for this singular behaviour. Since using sul- 
 phurous acid only, he has never been troubled with any- 
 thing similar. 
 
 It is exceedingly difficult to wash out the chromic 
 oxide ; it succeeds best in the following way : After the 
 precipitate has settled, the clear liquid is passed through 
 the filter, then boiling water is added to the precipitate, 
 and, after settling, the supernatant liquid is filtered ; the 
 precipitate then is put on the filter, and washed twice or 
 three times with boiling water ; it is then washed back 
 again into the dish and boiled with water until the little 
 lumps which clog together are completely broken up, and 
 it is then filtered again, and this operation repeated until 
 the wash- waters do not show the presence of any sul- 
 phates when tested with barium chloride. The precipi- 
 tate is then dried and burned. No matter how well it 
 may have been washed, it almost invariably contains 
 minute quantities of alkalies, in the presence of which a 
 little chromic oxide is converted into chromic acid. The 
 ignited precipitate is therefore put into a dish, boiled 
 with water, a few drops of sulphurous acid added, precipi- 
 tated by ammonia, filtered, washed, dried, ignited, and 
 weighed. 
 
 In this manner the chromic oxide is obtained quite 
 pure, and repeated analyses of the same sample of ore 
 never vary 0-25 per cent, of chromic acid. 
 
 Mr. O'Neill uses a volumetric method to estimate the 
 chromic acid, depending upon the capability of sulphurous 
 
WOLCOTT GIBBS'S PEOCESS. 823 
 
 acid to deoxidise chromic acid at the ordinary temperature 
 in the presence of free sulphuric acid. Prepare a strong 
 solution of sodium bisulphite, by passing sulphurous acid 
 through caustic soda to saturation, and then make it 
 alkaline with caustic soda, so as to have a neutral sulphite, 
 which is less readily oxidised by keeping than the bisulphite. 
 Use a dilute solution made from this concentrated sulphite^ 
 of such a strength that one grain of pure potassium bichro^ 
 mate requires about, and not less than, 200 grains measure 
 of the sulphite to deoxidise it. The value of the sulphite 
 must be estimated for every operation, since it is con- 
 tinually absorbing oxygen. This is done twice by weighing 
 out three grains, and four grains of pure bichromate, dis- 
 solving each of them in ten ounces of water and acidulating 
 freely with sulphuric acid, then adding the sulphite from 
 the burette, with continual stirring, until the chromic acid 
 is destroyed. The stopping-point may be ascertained by 
 the colour when one is accustomed to the reaction, but 
 even an experienced eye will often be glad of additional 
 evidence. A mixture of potassium iodide and boiled 
 starch, slightly acidulated, forms a delicate test ; it has 
 usually a faint colour, which is even preferable to a colour- 
 less mixture. An exceedingly small quantity of chromic 
 acid develops the blue colour in spots of this mixture, 
 and a very slight excess of sulphite makes it colourless. 
 One division of the sulphite test-liquor, or 0-05 grain of 
 potassium bichromate in twelve ounces of water, easily 
 and quickly influences the test mixture. The chr ornate 
 from the chrome ore is tested in the same manner, and the 
 quantity of chromium oxide or chromic acid calculated 
 from the equivalents of potassium bichromate. A tenth 
 of a grain more than five is taken to allow for all losses, 
 and the results are multiplied by 20 for the percentage. 
 151 of potassium bichromate is reckoned equivalent to 80 
 of green oxide of chromium, and 104 of chromic acid. 
 An estimation can be made by this process in three 
 or four hours, and a double estimation in a little longer 
 time. 
 
 Dr. Wolcott Gibbs has shown that chrome-iron ore 
 
824 THE ASSAY OF CHROMIUM. 
 
 may be completely resolved by fusion with fluohydrate of 
 potassium fluoride. In this and in all similar applications 
 of the fluohydrate it is best to evaporate the finely pul- 
 verised mineral to dryness with a concentrated solution of 
 the salt. On subsequently heating to low redness, the 
 resolution of the mineral is effected with the utmost ease, 
 a portion of the chromium being usually oxidised to 
 chromic acid by the oxygen of the air. After expelling 
 the fluorine by heating the fused mass with sulphuric acid, 
 the remaining mass is dissolved in water, rendered nearly 
 neutral by a solution of sodium carbonate, and sodium 
 acetate is added in excess. A current of chlorine gas, or 
 a solution of chlorine water, then readily converts the 
 whole of the chromium present into chromic acid, espe- 
 cially when the solution is hot, and when it is kept nearly 
 neutral by the occasional addition of sodium carbonate. 
 The excess of chlorine is easily got rid of by boiling. 
 
 The iron and alumina may then be precipitated to- 
 gether by boiling the solution in the presence of excess of 
 sodium acetate. It is more convenient and equally accu- 
 rate to neutralise the solution with ammonia, separate the 
 ferric oxide by filtration, and estimate the chromium in 
 the filtrate by reduction and precipitation with ammonia. 
 
 For the technical estimation of chromium in chro- 
 mite, Mr. Clarke fuses with cryolite and potassium bisul- 
 phate. The mass is then treated with a little strong 
 hydrochloric acid, and allowed to digest for about ten 
 minutes. Then upon boiling with water the whole dis- 
 solves. The solution should then be neutralised, sodium 
 acetate added, and the chromium oxidised to chromic acid 
 by a current of chlorine gas, or by boiling with sodium 
 hypochlorite solution. The chromium may then be sepa- 
 rated from other substances, as directed by Dr. W. Gibbs. 
 When chromite is fused with potassium bisulphate and 
 cryolite, and saltpetre is added to the mass, as soon as clear 
 fusion is obtained, the chromium is nearly all oxidised to 
 chromic acid. If the mass be boiled with a solution of 
 sodium carbonate, and the liquid filtered, a filtrate is 
 obtained which contains nearly all, but not quite all, the 
 chromium as alkaline chromates, free from iron or alumina ; 
 
ESTIMATION OF CHROMIUM IN CHROME IRON ORE. 825 
 
 but invariably the residue upon the filter contains traces 
 of chromium. When chromite is fused with the acid 
 potassium fluoride, a part of the chromium is usually 
 oxidised to chromic acid by the oxygen of the air ; and in 
 one case that came under Mr. Clarke's observation, when 
 he heated the resulting mass with sulphuric acid, red fumes 
 were given off, which were probably the so-called chro- 
 mium terfluoride. 
 
 When potassium bisulphate alone is used for the de- 
 composition of chromite, &c., it is necessary that the 
 mineral should be reduced to extremely fine powder ; but 
 when the mixture of bisulphate and fluoride is employed, 
 although the mineral should be in fine powder, such an 
 extreme state of subdivision is by no means required, and 
 thus much labour is saved. 
 
 Estimation of Chromium in Chrome Iron Ore. When 
 an estimation of the chromium only is sought, the decom- 
 position of chrome iron ore, according to H. N. Morse 
 and W. C. Day, can be best accomplished by fusing the 
 material with potassium hydroxide in a wrought-iron 
 crucible. 
 
 The method here described has, without exception, 
 given satisfactory results. From 6 to 10 grms. of potas- 
 sium hydroxide are placed in a wrought-iron crucible 
 (having the form of the ordinary porcelain crucible and a 
 capacity of about 100 c.c.) and gently heated until the 
 evolution of steam ceases and the fused mass becomes 
 tranquil. After cooling, the finely pulverised material, 
 weighing not more than 0-5 grm., is placed upon the 
 potassium hydroxide and evenly distributed over the 
 surface. A flame just sufficient to thoroughly fuse the 
 alkali is applied to the uncovered crucible, and the con- 
 tents, as long as they remain in a fluid condition, fre- 
 quently stirred with a piece of iron wire, which is allowed 
 to remain in the crucible. The decomposition progresses 
 rapidly, and the potassium hydroxide together with the 
 soluble products of the decomposition soon begins to rise 
 upon the sides of the crucible, where it deposits itself in 
 forms somewhat resembling the cauliflower. Within two 
 or three hours the decomposition is complete, and the 
 
826 THE ASSAY OF CHEOMIUM. 
 
 bottom of the crucible becomes dry. The crucible is 
 then turned upon its side and the temperature of its under 
 surface raised to a dull red heat. The incrustation on the 
 interior of the crucible does not fuse at this temperature, 
 but becomes rapidly yellow, owing to the oxidation of 
 the chromium to chr ornate. At the end of two or three 
 hours the oxidation is perfect. Portions of the incrusta- 
 tion retain a greenish colour, however long the heating is 
 continued ; but this is due to the presence of iron or 
 manganese, and not to unoxidised chromium. 
 
 After cooling, the crucible is placed in a porcelain 
 evaporating dish and the contents removed by means of 
 hot water. The solution, which at first has a greenish 
 appearance owing to the presence of iron dissolved in the 
 caustic potash, is heated for some time in order to effect 
 complete precipitation of the iron. The filtrate, which 
 has a clear yellow colour, is rendered slightly acid with 
 pure dilute nitric acid, the aluminium precipitated with 
 ammonia and washed by decantation. The potassium 
 chromate is then reduced and the silica rendered insoluble 
 by evaporating to perfect dryness with an excess of hydro- 
 chloric acid. The residue is moistened with hydrochloric 
 acid and treated with water. 
 
 It only remains to separate the chromium in the 
 filtrate from magnesium and to estimate it as chromic 
 oxide. To do this the authors prefer in each instance to 
 first precipitate with barium carbonate. 
 
 We give below the data of twelve estimations upon 
 material whose origin is unknown to us. The analyses 
 reported were in each case made consecutively. 
 
 Wt. of Ore Taken. Wt. of Cr 2 3 Found. Percentage. 
 
 1. 0-3250 0-1343 41-32 
 
 2. 0-3465 0-1433 41-35 
 
 3. 0-3248 0-1343 41-36 
 
 4. 0-3383 0-1392 41-14 
 
 5. 0-2723 0-1120 41-12 
 
 6. 0-3955 0-1629 41-19 
 
 7. 0-3311 0-1358 41-02 
 
 8. 0-31285 0-1290 41-23 
 
 9. 0-3387 0-1392 41-10 
 
 10. 0-2995 0-1235 41-25 
 
 11. 0-3606 0-1478 . 41-00 
 
 12. 0-3007 0-1236 41-11 
 
ESTIMATION OF CHROMIUM IN IRON AND STEEL. 827 
 
 Estimation of Chromium by means of Standard Solu- 
 tion. This process is the converse of the estimation of iron 
 by means of solution of potassium chromate. 
 
 The chrome ore is treated with potassium nitrate and 
 sodium carbonate, as above described ; and the solution of 
 potassium chromate so obtained has an excess of hydro- 
 chloric acid added to it. 
 
 It is stated at page 324, under the head of Assay of 
 Iron in the Wet Way, that 100 parts of metallic iron cor- 
 respond to and are represented by 88-6 grains of potassium 
 bichromate. Now 88-6 grains of potassium bichromate 
 contain 32-96 grains of chromium ; therefore 100 grains of 
 iron are equal to 32-96 of chromium. From these data a 
 standard solution can be readily made : thus Dissolve 50 
 grains of pianoforte wire in excess of hydrochloric acid ; 
 place the solution in a burette, and fill up to 100 on the 
 instrument with water, and well mix : it is now evident 
 that every division of the burette will equal or represent 
 0-1648 grain of chromium. The assay is now thus proceeded 
 with : Gradually add the standard solution of iron to the 
 solution of potassium bichromate acidulated with hydro- 
 chloric acid, until a drop of the solution mixed with a drop 
 of solution of potassium ferrocyanide gives a pale blue 
 colour : a slight excess of ferrous oxide is then present, 
 showing that all the chromic acid has been reduced to the 
 state of chromium oxide. Now observe how many divi- 
 sions of the iron solution have been required, and multiply 
 them by -1648 : the resulting number will represent the 
 amount of metallic chromium in the sample submitted to 
 assay. 
 
 For the estimation of chromium in iron and steel 
 Mr. J. 0. Arnold (' Chem. News,' Dec. 10, 1880, p. 285) 
 weighs out from 1 to 5 grms. of steel (in drillings) accord- 
 ing to the amount of chromium present (this may be 
 ascertained by a rough colorimetric test). Place the 
 metal in a wide, covered beaker, and add 20 c.c. of strong 
 hydrochloric acid ; heat till all action is at an end ; rinse 
 the cover and sides of the beaker from splashings, and eva- 
 porate the solution gently to complete dryness. If the 
 
g28 THE ASSAY OF CHEOMIUM. 
 
 evaporation has not been too rapid, the chlorides may be 
 almost entirely detached from the bottom of the beaker in 
 a brittle cake. This is broken up into small pieces by means 
 of a platinum spatula, and as much as possible is brushed 
 out into a clean dry porcelain dish. A small quantity of 
 the chlorides will, however, remain adhering to the beaker : 
 this may be removed with 2 or 3 c.c. of dilute hydro- 
 chloric acid. The solution is poured into a deep platinum 
 crucible, the beaker rinsed with 1 or 2 c.c. of water, the 
 washings added to the crucible, the contents of which are 
 now evaporated to dryness on a sand-bath. When dry 
 the main quantity of the chlorides is carefully brushed out 
 of the porcelain into the platinum dish, and is reduced to 
 a fine powder by means of a little pestle made from a glass 
 rod. The spatula and pestle are cleaned into the crucible. 
 The finely divided chlorides are now thoroughly mixed 
 with an excess of fusion mixture (1 part dry sodium car- 
 bonate to 1 part powdered nitre), a cover is placed over 
 the crucible, and its contents are fused over a gas blow- 
 pipe till quite liquid. By this treatment the iron is con- 
 verted into insoluble oxide, the manganese, silica, and 
 chromium respectively into alkaline manganate, silicate, 
 and chromate. When cool the crucible is placed in a 
 beaker containing 80 c.c. of boiling water, and is gently 
 boiled till the fused mass is detached and dissolved out. 
 This may be assisted by occasional stirring with a glass 
 rod. When clear from oxide, remove the crucible and 
 cover, washing them well with hot water. Add 3 or 4 
 drops of alcohol to decompose the manganate, and allow 
 the iron and manganese oxides to settle thoroughly. 
 When the supernatant liquid is quite clear it is decanted 
 on a close double filter, the filtrate being received into a 
 clean beaker. The precipitates are disturbed as little as 
 possible. When all the clear liquid has passed through, the 
 filter is well washed with hot water. The precipitates are 
 now washed twice by decantation with 30 c.c. of hot water ; 
 at the second washing the contents of the beaker are allowed 
 to drain gently into the filter, which is now allowed to 
 drain thoroughly, and is removed without further washing. 
 
VOLUMETRIC ESTIMATION OF CHROMIUM. 829 
 
 These precautions in washing must be strictly carried out, 
 as the ferric oxide is in such an exceedingly fine state of 
 division that any attempt to wash or disturb it on the 
 filter will inevitably cause some of it to pass through into 
 the chromium solution. The clear yellow filtrate contains 
 chromium and silica, to it is added 20 c.c. of hydrochloric 
 acid, the cover being kept on the beaker, to prevent the 
 projection of the solution by the evolved carbonic acid. 
 The solution is now well boiled until all the carbonic acid 
 and nitrous fumes are driven off. Its colour will now be 
 green, owing to reduction to chloride. Dilute ammonia 
 is added until alkaline, and the solution heated nearly to 
 boiling. The resulting precipitates consist of chromium 
 silicate mixed with alkaline salts. It is collected on a 
 filter (previously well washed with hot dilute hydrochloric 
 acid to free it from iron), and is slightly washed. When 
 the washings have drained through, the precipitate is dis- 
 solved off the filter with hot dilute hydrochloric acid, the 
 filtrate being received into the beaker in which the precipi- 
 tation took place. The solution is now evaporated to dry- 
 ness to relider the silica insoluble. The soluble portion is 
 taken up with 10 c.c. hydrochloric acid and 90 c.c. of water, 
 and is filtered through a washed filter into a clean beaker, 
 the filter being well washed. The precipitation is now 
 repeated as above, and the chromic hydrate comes down 
 free from silica and alkaline salts. It is collected, washed, 
 dried, ignited, and weighed as chromic oxide. 
 
 Precautions. Only three or four drops of alcohol 
 should be added, as this quantity is sufficient, not only to 
 precipitate the manganese, but also to effect the reduction 
 to chloride. If too much is added organic compounds are 
 formed, which tend to prevent the complete precipitation 
 of the hydrate. The ammonia in the last precipitation 
 should be added in the least possible excess, and the solu- 
 tion should be heated gently nearly to boiling. If any 
 great excess of ammonia be present, and if the solution is 
 boiled, the glass of the beaker is attacked and the result is 
 high. The method, if properly carried out, is accurate. 
 
 Mr. W. J. Sell has devised the following method for 
 
830 THE ASSAY OF CHROMIUM. 
 
 the volumetric estimation of chromium. The solution, 
 containing chromium acidified with sulphuric acid, is 
 boiled, and a dilute solution of permanganate added to the 
 boiling liquid until a purplish tint remains after boiling for 
 three minutes. The solution is then rendered slightly 
 alkaline with sodium carbonate, alcohol is added, and the 
 manganese filtered off. The chromic acid in the filtrate is 
 estimated, by titration with iodine and sodium hyposul- 
 phate. The author has successfully applied the method 
 to the estimation of chromium in chromic iron ore. He re- 
 commends the following plan for effecting its decomposition. 
 The chromic iron ore is placed on the top of about ten times 
 its weight of a mixture, made in the proportion of one 
 molecule of well-fused and powdered sodium bisulphate 
 to two molecules of sodium fluoride, and the whole is 
 ignited for fifteen minutes. An amount of sodium bisul- 
 phate is now added equal to that of the mixture taken, 
 and when thoroughly fused a further addition of an equal 
 quantity of bisulphate is made, the mass fused, and then 
 rapidly cooled. The fused mass so obtained dissolves 
 completely in boiling water acidified with sulphuric acid. 
 In this way an estimation can be made in an hour and a 
 quarter. 
 
831 
 
 CHAPTER XXI. 
 
 THE ASSAY OF AKSENIC. 
 
 THE minerals from which arsenic is produced are the 
 following : 
 
 Native arsenic. 
 
 Arsenical pyrites, FeS 2 + FeAs, containing 46-6 As and 19-6 S. 
 
 Arsenical pyrites, Fe 4 As 3 , containing 66'8 As. 
 
 Speiskobalt (Co,NiFe), As. 
 
 Glanzkobalt, CoS 2 + CoAs, 
 
 Coppernickel, NijAs. 
 
 Nickel and cobalt arsenical pyrites (Co,Ni,Fe)S 2 + (Co,Ni,Fe), As, 
 
 White arsenical nickel, NiAs; Tennantite (Cu2S,SeS) 4 ,AsS 3 . 
 
 Eealgar, AsS 2 and yellow arsenic AsS 3 . 
 
 Assay of Arsenic. 50 grains of the finely pulverised 
 mineral are deflagrated with 200 of potassium nitrate and 
 200 of sodium carbonate in a porcelain crucible. When 
 the crucible is cold, it and its contents are to be treated 
 with water, as in the case of chromium. The solution will 
 contain potassium arseniate, and (if the ore had in its con- 
 stitution sulphur, which is most likely) potassium sulphate. 
 Lead nitrate must be added to the solution (made neutral 
 with nitric acid, if requisite) : a mixture of lead arseniate 
 and sulphate is precipitated ; this precipitate is well washed 
 on a filter, and digested with dilute nitric, acid ; this agent 
 dissolves out the lead arseniate, and leaves the sulphate. 
 Filter, and saturate the filtered solution with soda, which 
 will throw down the arseniate ; this must be collected on 
 a filter, washed, dried, and weighed. Every 100 parts cor- 
 respond to 22-2 of metallic arsenic, or 29 parts of arsenious 
 acid (the white arsenic of commerce). 
 
 This method is only approximative : the following is 
 the better plan : 
 
 Digest the ore in strong nitric acid until nothing more is 
 
832 THE ASSAY OF AESENIC. 
 
 taken up (the action may be facilitated by the occasional 
 addition of a crystal or two of potassium chlorate), and all 
 action on the addition of fresh acid is at an end : dilute 
 with water, and filter : to the filtered solution add lead 
 nitrate, and proceed as above. 
 
 For estimating the amount of arsenic in ores, Mr. 
 Parnell says that the neatest, simplest, and most accurate 
 mode of procedure is to heat the finely divided sample in 
 a gentle stream of chlorine gas to a temperature of about 
 200 C., and to collect the escaping arsenic chloride in 
 chlorine-water. If free from antimony, the liquid may be 
 well boiled, to expel free chlorine, and the arsenic preci- 
 pitated with sulphuretted hydrogen, and weighed as penta- 
 sulphide. 
 
 In cases where the arsenic is obtained in the form of 
 arsenio-magnesian phosphate (as in the separation of the 
 metal from antimony or copper), the most accurate plan 
 would be to dissolve the precipitate in hydrochloric acid, 
 and precipitate the arsenic as pentasulphide. When the 
 amount of arsenic is small, it may be weighed as the 
 double arseniate. The sample should not, however, be 
 dried at a higher temperature than that of an ordinary 
 water-bath namely, about 95 C. Perfectly accurate 
 results could, no doubt, be obtained by drying the pre- 
 cipitate over sulphuric acid, when it retains its six equi- 
 valents of water. The only objection is that it would 
 take many days for a filter containing a precipitate to be 
 properly dried by this means. 
 
833 
 
 ''*)) 
 
 CHAPTER XXII. 
 
 THE ASSAY OP MANGANESE. 
 
 THE following are the commercially valuable minerals 
 containing manganese. 
 
 Pyrolusite, MnO 2 , containing 18-0 p.c. of available oxygen 
 
 Braunite, Mn 2 3 , 10-0 
 
 Manganite, Mn 2 O 3 , 9'0 
 
 Varvicite, Mn" 2 O 2 + Mn,0 3 ,HO, 13-8 
 
 Hausmannite, MnO + Mn 2 6 3 , 6*8 
 
 Psilomelane, Mn 2 O 3 . 
 
 The assay of this metal is confined to the amount of 
 peroxide any one of its ores may contain. There are 
 several methods of effecting this, and the best of these will 
 be described below. 
 
 Valuation of Manganese Ores. The best methods used 
 for the valuation of manganese ores are not necessarily 
 those which give in the most rapid and accurate manner 
 the absolute amount of manganese peroxide present in the 
 ore. The analyst must bear in mind that the commercial 
 value of manganese ore depends on its power of liberating 
 chlorine from hydrochloric acid ; and it not unfrequently 
 happens that an ore, which on accurate analysis would be 
 reported to contain a high percentage of manganese per- 
 oxide, likewise contains some other mineral (iron protoxide 
 or magnetic oxide), which will materially reduce the value 
 of the manganese as a chlorine-yielding ore. It is on this 
 account that some processes excellent though they be 
 from a purely analytical point of view have fallen into 
 discredit amongst manufacturers, whilst other processes 
 which do not profess to give the amount of manganese per- 
 oxide actually present, but only that available for liberating 
 chlorine, are now generally adopted. In the following 
 
 3H 
 
834 THE ASSAY OF MANGANESE. 
 
 pages are given the methods of testing manganese ore for 
 the available peroxide which have best stood the test of 
 practical experience. 
 
 Messrs. Scherer and Kumpf, after examining all the most 
 approved methods in Dr. Fresenius's laboratory at Weis- 
 baden, have come to the conclusion that Bunsen's method 
 is the best for rapidly giving the amount of available man- 
 ganese in an ore. This process is carried out by dissolving a 
 weighed quantity of the sample in strong hydrochloric acid 
 in a small flask, until complete decomposition has taken 
 place. The escaping chlorine is received in a strong solution 
 of potassium iodide, and the liberated iodine subsequently 
 estimated by means of a standard solution of sodium hypo- 
 sulphite and a solution of starch. To prevent the solution 
 of potassium iodide from being sucked back into the gene- 
 rating-flask, a few small pieces of magnesite are introduced 
 with the manganese, so that a continual slight escape of 
 carbonic acid takes place through the solution. The solu- 
 tion of sodium hyposulphite is tested by means of carefully 
 prepared pure iodine, dissolved in potassium iodide. The 
 solution should be of such a strength that 1000 c.c. of 
 sodium hyposulphite solution corresponds to from 2 to 
 3 grms. of manganese peroxide. In this estimation the 
 iodine liberated by the chlorine should be tested as soon 
 as possible after the decomposition ; it gives higher results 
 after standing 24 hours than before. These higher results 
 are caused by the liberation of iodine by spontaneous 
 decomposition of hydriodic acid, set free by the hydro- 
 chloric acid, distilled over during the process. The follow- 
 ing experiment proves this : A few drops of hydrochloric 
 acid were added to a solution of potassium iodide. The solu- 
 tion remained for some hours colourless, but, after standing 
 twenty-four hours, had become quite yellow, and was found 
 to contain free iodine sufficient to indicate 8 per cent, of 
 manganese peroxide when titrated with hyposulphite. 
 
 Messrs. Scherer and Eumpf have made the suggestion 
 that the value of manganese ores should be measured by 
 chlorometrical degrees rather than by the actual percentage 
 of binoxide ; thus tending in the same direction as the 
 
VALUATION OF MANGANESE ORES. 835 
 
 resolution* passed by the Association of Alkali Manu- 
 facturers in 1869, in reference to this subject a decision 
 which would seem also "to indicate a desire on the part of 
 manufacturers that tests of manganese ore should express 
 the amount of peroxide available for liberating chlorine, 
 and not the amount actually present in the ores. 
 
 For the above reasons, Dr. Paul adopts Mohr's method 
 of using a known quantity of a standard solution of oxalic 
 -acid, together with excess of sulphuric acid, for dissolving 
 the ore ; if necessary, boiling until the ore is completely 
 dissolved and then, by means of a standard solution of 
 permanganate, estimating the quantity of oxalic acid 
 remaining undecomposed. This method is very conve- 
 nient for testing manganese ores, and involves only one 
 weighing for each test. The results obtained are also 
 very uniform. 
 
 This method has also the advantage of giving results 
 which fairly represent the amount of available peroxide 
 in manganese ores ; for any iron that may be present as 
 metal or protoxide would consume an equivalent quantity 
 of permanganate solution, and thus apparently reduce the 
 quantity of oxalic acid decomposed by the peroxide to 
 an extent proportionate to the amount of iron existing in 
 the ore. Thus, for instance, if the quantity of oxalic acid 
 decomposed by 100 grains of manganese ore free from 
 iron or ferrous oxide were 109;53 grains, the ore would 
 contain 76*5 per cent, of peroxide, and the whole of that 
 would be available. But if the 100 grains of ore also 
 contained 5*6 grains of metallic iron, or an equivalent of 
 protoxide, the permanganate solution required for per- 
 oxidising that iron would represent 6-3 grains of oxalic 
 acid, and the quantity of oxalic acid decomposed by the 
 peroxide would appear so much less than it really was, or 
 103-23 grains instead of 109-53 grains. Accordingly, the 
 amount of peroxide would be represented as 72-1 per 
 cent, instead of 76-5 per cent. : and that would, in fact, 
 
 * ' That as the testing of manganese according to the method of Will and 
 Fresensius is, in the opinion of the meeting, incorrect, and yields uncertain 
 results, it is recommended to members of this association not to buy by that test.' 
 
 3 H 2 
 
836 THE ASSAY OF MANGANESE. 
 
 be the amount of peroxide available for generating 
 chlorine. 
 
 This method of testing recommends itself by its sim- 
 plicity, and by the fact that the standard solutions of 
 oxalic acid and permanganate will keep for a long time 
 without alteration of value. The oxalic acid solution 
 contains 63 grms. in the litre, and 1 c.c. is equivalent to 5 
 c.c. of the permanganate solution. 
 
 Mr. John Pattinson prefers a modification of Otto's 
 process for the valuation of manganese, which consists in 
 boiling the ore with a known weight of a proto-salt of iron, 
 and then estimating the excess of iron by a standard 
 solution of potassium bichromate. This modified process, 
 in his opinion, requires less skill and care at the hands 
 of the operator than Bunsen's method, as described by 
 Messrs. Scherer and Eumpf. 30 grs. of clean iron wire 
 are placed in a 20-oz. flask, along with 3 oz. of dilute 
 sulphuric acid, made by adding 3 parts of water to 1 of 
 oil of vitriol. A cork through which passes a tube bent 
 twice at right angles is inserted in the neck of the flask, 
 and the flask is heated over a gas-flame until the iron is 
 dissolved. The bent tube is placed so as to dip into a 
 small flask or beaker containing a little water. When the 
 iron is quite dissolved, 30 grs. of the finely pounded and 
 dried sample of manganese ore to be tested are put into 
 the flask, the cork replaced, and the contents again made 
 to boil gently over a gas-flame until it is seen that the 
 whole of the black part of the manganese is dissolved. 
 The water in the small flask or beaker is then allowed to 
 recede through the bent tube into the larger flask, more 
 distilled water is added to rinse out the small flask or 
 beaker and bent tube, the cork well rinsed, and the con- 
 tents of the flask made up to about 8 or 10 oz. with dis- 
 tilled water. The amount of iron remaining unoxidised 
 in the solution is then ascertained by means of a standard 
 solution of potassium bichromate. The amount the bi- 
 chromate indicates, deducted from the total amount of 
 iron used, gives the amount of iron which has been 
 peroxidised by the manganese ore, and from this can be 
 
VALUATION OF MANGANESE ORES, 837 
 
 calculated the percentage of manganese peroxide contained 
 in the ore. Thus, supposing it were found that 4 grs. of 
 iron remained unoxidised, then 30 4=26 grs. of iron, 
 which have been oxidised by the 30 grs. of ore. By a 
 simple calculation it is found that this 26 grs. of iron 
 are equivalent to 20-43 grs. of manganese peroxide, the 
 amount of peroxide in the 30 grs. of ore. The percent- 
 age is, therefore, 68*10. 
 
 It must be understood that an ore may contain a mix- 
 ture of one atom of manganic oxide, and two atoms of 
 magnetic iron oxide, or 27'3 per cent, of the former, with 
 72 -7 per cent, of the latter ; in such a mixture the method 
 of Fresenius and Will will indicate with precision the 
 amount of manganese peroxide, but on adding hydrochloric 
 acid to this mixture not a trace of chlorine will be given 
 off, since the free atom of oxygen of the manganese 
 peroxide is just sufficient for the oxidation of the 2 atoms 
 of iron protoxide of the magnetic iron ore ; in the same 
 way a mixture of manganese binoxide with iron protosul- 
 phate or protocarbonate of that metal will be perfectly 
 worthless as an article for chlorine-making use. 
 
 Dr. Mohr accordingly recommends that manganese 
 ores and samples of manganese peroxide should be 
 always tested, previous to analysis, with an astatic mag- 
 netic needle, and he further recommends Dr. Bunsen's 
 process (given on p. 834) as the best and surest method 
 of analysis. This process is really the same as that which 
 the manufacturer employs for making chlorine ; any 
 magnetic iron ore present will become oxidised in both 
 processes, and a special examination for magnetic iron 
 oxide is rendered unnecessary, while the available manga- 
 nese for the production of chlorine only is estimated. 
 
 Mr. J. Pattinson proposes the following method of 
 precipitating manganese entirely as peroxide, and applies 
 it to the volumetric estimation of manganese. He 
 finds that the whole of the manganese in a solution of 
 manganous chloride can be precipitated as peroxide, if a 
 certain amount of ferric chloride be present, by a suf- 
 ficient excess of a solution of chloride of lime or bromine 
 
838 THE ASSAY OF MANGANESE. 
 
 water, adding, after heating the solution to from 140 to- 
 160 F., an excess of calcium carbonate, and then well 
 stirring the mixture. Zinc chloride may be substituted 
 for ferric chloride. The author recommends the follow- 
 ing solutions, etc. The clear liquid obtained by decanta- 
 tion from a 1*5 per cent, solution of bleaching-powder ; 
 light granular calcium carbonate obtained by precipitating 
 an excess of calcium chloride by sodium carbonate at 
 180 F. ; a 1 per cent, solution of ferrous sulphate in 
 dilute (1 in 4) sulphuric acid ; standard solution of potas- 
 sium dichromate equivalent to 1 part of iron in 100 of 
 solution. The application of the process to manganiferous 
 iron ores is as follows : 10 grains of the ore, dried at 
 212, are dissolved in a 20-oz. beaker in about 100 fluid 
 grains of hydrochloric acid (sp. gr. 1*18). Calcium 
 carbonate is then added, until the free acid is neutralised 
 and the liquid turns slightly reddish. 6 or 7 drops of 
 hydrochloric acid are now added, and 1000 grains of 
 the bleaching-powder solution, or 500 grains of saturated 
 bromine water, and boiling water run in until the tempe- 
 rature is raised from 140 to 160 F. ; 25 grains of calcium 
 carbonate are added, and the whole well stirred. If the 
 supernatant solution has a pink colour, the permanganate 
 is reduced by a few drops of alcohol. The precipitated 
 oxides of iron and manganese are filtered off and washed. 
 1000 grains of the acidified ferrous sulphate solution are 
 carefully measured into the 20-oz. beaker already used ; 
 the filter, with its washed contents, added. A certain quan- 
 tity of the ferrous sulphate is oxidised by the manganese* 
 binoxide ; this quantity is estimated with the standard 
 dichromate solution, when the quantity of manganese bin- 
 oxide can easily be calculated. The iron present must be 
 at least equal in weight to the manganese during the pre- 
 cipitation, in order to insure the absence of lower oxides. 
 
839 
 
 CHAPTEE XXIII. 
 
 THE ASSAY OF NICKEL AND COBALT ORES. 
 
 ORES OF NICKEL. 
 
 Nickel oxide. 
 
 Nickel sulphide. 
 
 Nickel arsenide ; kupfernickel. 
 
 Nickel arsenio-sulphide ; grey nickel. 
 
 Nickel antimonio-sulphide. 
 
 Nickel arseniate. 
 
 Nickel arsenite. 
 
 Nickel silicate. 
 
 ORES OF COBALT. 
 
 Cobalt oxide. 
 
 Cobalt sulphide ; cobaltine. 
 
 Cobalt sulphate. 
 
 The cobalt arsenides. 
 
 Arsenio-sulphide, or grey cobalt. 
 
 Cobalt arsenite. 
 
 THE analysis of cobalt ores is the most tedious, with the 
 exception of those of platinum, of any that fall under the 
 assayer's notice, the greatest difficulty being in the sepa- 
 ration of cobalt and nickel. The following is Mr. Hadow's 
 process : 
 
 The only examination which the ore need undergo 
 previously to the solution of a weighed quantity is with the 
 view of obtaining a rough idea as to the amount of arsenic 
 and cobalt or nickel present in the sample ; for this purpose 
 a little may be roasted on charcoal, or ignited in a tube? 
 to see whether arsenic readily sublimes ; another portion, 
 of a few grains' weight, may be dissolved in aqua regia in a 
 test-tube, when the depth of the blue or green colour will 
 serve as an indication of the degree of richness of the ore 
 in cobalt and nickel. 
 
 If the ore is rich, from 20 to 30 grains ; if poor, from 
 
840 THE ASSAY OF NICKEL AKD COBALT. 
 
 50 to 100 grains, in a state of fine division, are weighed 
 out for the analysis. If much arsenic has been found, the 
 portion, after weighing, had better be ignited in a small 
 porcelain capsule or crucible over a gauze burner, when it 
 generally ignites and smoulders away, evolving abundance 
 of arsenious acid. The powder ready for solution is trans- 
 ferred to a small 4-oz. flask by means of glazed letter-paper 
 and a camel's-hair paint-brush to sweep in the last par- 
 ticles ; the mouth of the flask is then partially closed by a 
 small funnel placed to catch the drops projected during 
 solution. The ore is then drenched with hydrochloric 
 acid, nitric acid being added from time to time, until all 
 heavy metallic- looking particles are found to have disap- 
 peared from the bottom of the flask. The solution may 
 then be decanted from the insoluble matters into a half- 
 pint beaker, together with the washings of the flask ; and 
 as sulphur frequently remains, entangling portions of undis- 
 solved ore, it is advisable to transfer the undissolved residue 
 fr6m the flask into a capsule, drying and igniting the con- 
 tents of the latter, and then digesting again the ignited 
 matters in a little more aqua regia ; the whole of the latter, 
 both dissolved and undissolved, may now be added to the 
 first portion in the half-pint beaker. 
 
 To separate out iron, arsenic, phosphoric acid, and 
 aluminium from the solution, sodium acetate may be added 
 at once, and the liquid boiled ; a far better mode, however, 
 is to effect a partial separation of these ingredients by the 
 addition of calcium carbonate in excess to the solution of 
 the ore, and after filtering out the solution containing the 
 greater portion of the cobalt and nickel, and partly wash- 
 ing the precipitate, to extract the last traces of cobalt and 
 nickel from the latter by dissolving it in hydrochloric acid, 
 adding excess of sodium acetate and boiling. The first 
 filtrate from the precipitate by calcium carbonate had 
 better be collected apart from the second filtrate from the 
 precipitate produced by sodium acetate, and received in a 
 beaker capable of holding at least a quart. The solution 
 of the precipitate by calcium carbonate is best effected in 
 a beaker, after the removal of the precipitate from the 
 
COBALT AND NICKEL ORES. 841 
 
 filter. This is easily effected by inclining the funnel over 
 the beaker and sending a stream of water from the wash- 
 bottle between the filter and the upper edge of the mass 
 of precipitate, when the latter will soon become detached 
 and slide off into the -beaker below : it is here treated with 
 dilute hydrochloric acid, to dissolve all but the insoluble 
 residues of the ore which had not been previously filtered 
 off, and then a solution of sodium acetate is added in 
 excess (indicated by the deep red colour of liquid) and the 
 whole, heated to boiling, may be filtered at once. Iron 
 thus separated out, in presence of free acetic acid, has less 
 tendency to retain cobalt than when precipitated by means 
 of calcium carbonate ; besides which the cobalt and nickel 
 in the filtrate are left in the condition of acetates, a 
 necessary step preparatory to their separation from man- 
 ganese, &c. 
 
 This method of separating out iron, &c., though very 
 effectual, was often at first found to be attended with diffi- 
 culties ; for if much arsenic were not present the basic iron 
 acetate frequently became slimy towards the end of the 
 filtration, only allowing the boiling washing-water to pass 
 with such extreme slowness as to render the method almost 
 useless, until it was found that the addition of a little 
 sodium sulphate during the washing at once and perma- 
 nently effected a cure, causing filtration to proceed rapidly, 
 and diminishing the tendency of the iron to pass the filter. 
 Another difficulty was, that when sodium acetate was added 
 at once to the original solution of the ore, the solution, 
 often containing much cobalt and nickel as acetates, and 
 filtered in a concentrated state, yielded to the filter-paper 
 sufficient cobalt and nickel to occasion distinct loss. This 
 was avoided by separating out the great bulk of the cobalt 
 and nickel in solution as chlorides by means of calcium 
 carbonate, as above recommended, and then the weaker 
 solution, being comparatively strongly acid, could be 
 filtered without loss. This second filtrate may still retain 
 traces of iron ; a little sodium acetate may be added to 
 make sure that none remains in the condition of chloride, 
 which would be indicated at once by a reddening of the 
 
842 THE ASSAY OF NICKEL AND COBALT. 
 
 liquid, and the whole is then boiled thoroughly once more ; 
 if rendered at all turbid passed, through a filter again, then 
 nearly neutralised with ammonia, and finally added to the 
 bulk of the cobalt and nickel solution in the quart beaker. 
 There will in all probability be enough of the sodium and 
 ammonium acetates present to convert the entire quantity 
 of cobalt and nickel into acetates without further addition, 
 and rendering it thus ready for the next operation. 
 
 If sulphuretted hydrogen be now transmitted through 
 the solution containing cobalt and nickel, these metals are 
 perfectly and completely separated without a trace of 
 manganese, magnesium, calcium, aluminium, or soluble 
 silica, which, when present, invariably accompany the sul- 
 phides precipitated by ammonium sulphide ; the sulphides, 
 moreover, thus precipitated from an acetic solution have 
 much less tendency to oxidise while on the filter, so that 
 their washing may be more perfectly accomplished than 
 in the former case. The passage of sulphuretted hydrogen 
 may be conveniently effected at the end of the day, and 
 the next morning the sulphides will be found perfectly 
 settled at the bottom of the beaker, permitting the great 
 bulk of the liquid (tested first to make sure of the removal 
 of cobalt and nickel) to be drawn off and thrown away, or 
 at least rapidly run through a filter ; the sulphides collected 
 at the bottom, together with that which always adheres to 
 the sides of the beaker, and which may be detached with- 
 out loss by a caoutchouc- covered glass rod, are then well 
 washed on the filter with boiling water until all soluble 
 matters are perfectly removed. The sulphides, perfectly 
 washed, are now to be dried by placing the funnel with 
 the filter in a broken beaker on wire gauze, at a safe 
 distance over a lamp, and when dry they may be detached 
 from the filter into a small beaker of from 1 to 2 oz. capa- 
 city, capable of being covered with a watch-glass ; the 
 filter itself is ignited, and the well-burnt ashes added to 
 the sulphides, which are then to be cautiously treated with 
 nitric acid, the action being rather violent, and, if care be 
 not taken, liable to occasion loss. With the aid of a little 
 heat, the whole should pass into solution. 
 
COBALT SPEISS. 84$ 
 
 In addition to cobalt and nickel the solution may still 
 contain zinc, together with copper, and other metals pre- 
 cipitable by sulphuretted hydrogen from hydrochloric 
 solutions. By passing sulphuretted hydrogen now through 
 the nitric solution, somewhat diluted, these latter are 
 readily precipitated and removed by filtration. Zinc, how- 
 ever, may still remain, to detect and remove which it is 
 necessary to expel the sulphuretted hydrogen still remain- 
 ing in the solution by boiling, to add solution of ammonia 
 until a precipitate occurs, and then to acidify pretty 
 strongly with acetic acid. If sulphuretted hydrogen slowly 
 transmitted, or fresh sulphuretted hydrogen water, occa- 
 sions a milkiness, zinc is present, and the slow passage of the 
 gas is to be continued until the precipitate begins to show 
 signs of darkening. The liquid is then filtered. The zinc 
 may be identified as such by collecting and igniting the 
 precipitate, when a trace of cobalt carried down with it 
 (and which may be separated out, if desired, by a repetition 
 of the process on the precipitate) will produce the beautiful 
 and well-known Einmann's green. 
 
 The filtrate, containing only nickel, cobalt, and salts 
 of ammonia, is treated with some pure sulphuric acid and 
 evaporated to dryness in a weighed capsule, and heated 
 sufficiently to expel the excess of sulphuric acid and all 
 the ammoniacal salts. The residual cobalt and nickel sul- 
 phates may now be weighed in a covered crucible. This 
 form of weighing these metals is easy, exact, and may be 
 rapidly executed. The weight of the ash of a filter of the 
 size used for collecting the sulphides must be ascertained 
 after treatment with sulphuric acid, and subsequent ex- 
 pulsion of the excess, and this weight deducted from the 
 total sulphates, in order to obtain perfectly correct results. 
 
 Decomposition of Cobalt Speiss or Arsenical Alloys of the 
 Same. The following is an outline of a new method de- 
 vised by Mr. H. Warren for decomposing cobalt speiss, or 
 arsenical alloys of the same : The regulus to be decom- 
 posed is first broken into pieces, about 1 Ib. in weight, and 
 suspended, by means of oiled string or other suitable sup- 
 port, in a vessel containing crude hydrochloric acid, to 
 
844 THE ASSAY OF NICKEL AND COBALT. 
 
 which has been added about 1 oz. of copper nitrate, and 
 the whole allowed to remain for the course of a day or so 
 to undergo dissolution, the copper nitrate reacting with 
 the hydrochloric acid present, forming cupric chloride, 
 while the nitric acid evolved reacts on the metals compos- 
 ing the regulus to form nitrates of the same, which in 
 their turn are again decomposed, generating by so doing 
 sufficient nitric acid to react on a further portion of un- 
 deconiposed regulus. 
 
 The alloy by this treatment is deprived largely of its 
 nickel and cobalt, besides other metals present, such as 
 arsenic and bismuth, which have passed into solution ; 
 while the remaining portion, having become sufficiently 
 brittle to be readily reduced to a powder, but still con- 
 taining notable quantities of both nickel and cobalt, is 
 calcined at a low red heat in a plentiful supply of air, by 
 which treatment the residue consisting of arsenides and 
 sulphides are wholly converted into arseniates and sul- 
 phates, which are readily brought into solution by means 
 of crude hydrochloric acid, and added to the original 
 solution. Metallic iron, in the form of bars, is now 
 brought into contact with the solution, by which means 
 the whole of the copper present is removed as metallic 
 copper, together with a considerable quantity of the arsenic 
 and bismuth present, the iron passing into solution as 
 chloride, which is again separated by the addition of milk 
 of lime, carrying with it the remainder of the arsenic as 
 basic arseniate of iron. The remaining salts of nickel and 
 cobalt still existing in solution are precipitated together 
 by means of sodium carbonate, as carbonate of nickel and 
 cobalt ; they are next disseminated through water, and 
 chlorine gas passed through until saturation, the whole of 
 the nickel by this means passing into solution, the cobalt 
 remaining precipitated. The solution containing the nickel 
 is brought to a state of ebullition, in order to free it from 
 the excess of chlorine present ; the nickel is precipitated as 
 hydrate, by means of caustic soda, ignited to expel the 
 water present, and reduced to the metallic state -by the 
 usual method. It has been found preferable, for various 
 
ASSAY OF METALLIC NICKEL. 845 
 
 reasons, to separate the calcium salts present, by means of 
 dilute sulphuric acid before proceeding to separate the 
 nickel and cobalt. 
 
 Assay of Nickel Ores. The ore of nickel usually met 
 with is the arsenide, containing variable quantities of 
 cobalt and iron, and frequently also copper and bismuth. 
 It is finely powdered, mixed with two parts of solid caustic 
 soda and 1^ part of sulphur, and fused in an earthen 
 crucible, gradually increasing the heat to dull redness, at 
 which temperature it is to be kept for some time. The 
 mass is then digested in water, which dissolves the soluble 
 sodium sulpho-arseniate, and leaves, when washed by de- 
 cantation, crystallised nickel sulphide. Attack this with 
 warm hydrochloric acid containing a little nitric acid. 
 
 The solution heated to about 70 C. is then submitted 
 to a current of sulphuretted hydrogen, which must be 
 continually passed until the liquid is cold. It is then to 
 be covered over and left at rest for 24 hours ; the arsenic 
 copper and bismuth come down as sulphides. These are 
 filtered off, and the filtrate is heated with potassium chlo- 
 rate or sodium hypochlorite to bring the iron to the state 
 of sesquioxide, which is then precipitated by ebullition 
 with excess of sodium acetate. The filtrate from the basic 
 iron acetate is concentrated by evaporation and mixed 
 with a saturated solution of potassium nitrite,* which will 
 precipitate all the cobalt. 
 
 The yellow precipitate washed with a saturated solu- 
 tion of potassium chloride may be treated as described 
 below. The nickel is precipitated from the liquid either 
 in the state of oxide by means of caustic potash, or after 
 concentration by a saturated and warm solution of potas- 
 sium binoxalate. The precipitated oxalate upon calcination 
 leaves pure nickel. 
 
 Assay of Commercial Metallic Nickel. Dissolve the 
 
 * Potassium nitrite is prepared by fusing, in an iron crucible, 1 part of 
 nitre with 2 parts of granulated lead, stirring well with an iron spatula, and 
 then heating to redness until the lead is completely oxidised. The fused mass, 
 after cooling, is extracted with water, and the small amount of lead which is 
 dissolved is precipitated by carefully adding a mixture of caustic ammonia 
 and ammonium carbonate or sulphide. 
 
846 THE ASSAY OF NICKEL AND COBALT. 
 
 metal in hydrochloric acid containing a little nitric acid. 
 Pass sulphuretted hydrogen through the solution until the 
 metallic impurities are thrown down, and then precipitate 
 the nickel and cobalt by a warm saturated solution of 
 potassium binoxalate. The precipitate after being washed 
 ,and calcined leaves the nickel (containing a little cobalt) 
 in the metallic state. 
 
 Assay of Cobalt Ores. Metallic cobalt may be prepared 
 from its ores (arsenide or sulphide) by a similar process 
 to that adopted in the case of nickel. When the mineral 
 contains more than 70 per cent, of arsenic, a preliminary 
 fusion should be performed with chloride of sodium, to 
 remove most of the arsenic. This may be continued by 
 roasting or by fusion with a mixture of sodium carbonate 
 and sulphur. 
 
 As nickel is almost invariably present in cobalt ores, 
 this metal will require to be separated. Wohler recom- 
 mends for this purpose the potassium nitrite process. The 
 yellow cobalt precipitate is dissolved in as small a quantity 
 of hydrochloric acid as possible, and sodium acetate is then 
 added ; the addition of a warm saturated solution of oxalic 
 acid now precipitates the cobalt as oxalate. This oxalate 
 after being washed and dried may be packed closely in a 
 crucible of biscuit porcelain, protected by enclosure in a 
 Hessian crucible. The covers being well luted on, the whole 
 is heated in a wind-furnace or a forge. If a sufficient tem- 
 perature has been obtained the cobalt will be in the form 
 .of a fused button. 
 
 Separation of Nickel and Cobalt. A method of sepa- 
 rating these metals, given some years since by Liebig, 
 consists in boiling the mixed double nickel and potassium 
 cyanides, cobalt and potassium cyanides, with oxide of 
 mercury. Nickel oxide is precipitated, while an equiva- 
 lent quantity of mercury is dissolved as cyanide. The 
 method certainly gives good results, but is not free from 
 objection. Long boiling is necessary before the precipi- 
 tation is complete, and it is difficult to prevent bumping 
 -during ebullition. The excess of mercury oxide must be 
 separated from the nickel oxide by a special operation, 
 
SEPARATION OF NICKEL AND COBALT. 847 
 
 and the nickel afterwards again precipitated by caustic 
 alkali. 
 
 According to Wolcott Gibbs,* these inconveniences 
 may be completely avoided by employing, instead of the 
 oxide alone, a solution of the oxide in the mercury 
 -cyanide. When this solution is added to a hot solution 
 of the double cyanide of nickel and potassium, the whole 
 of the nickel is immediately thrown down as a pale green 
 hydrate of the protoxide. Under the same circumstances 
 cobalt is not precipitated from the double cobalt and potas- 
 sium cyanide. Mr. W. N. Hill, who has repeatedly em- 
 ployed this method and carefully tested it, has found that 
 the separation effected is complete. JSFo cobalt can be 
 detected in the precipitated nickel oxide by the blow- 
 pipe, nor can the nickel be detected in the cobalt (finally 
 separated as oxide) by Plattner's process with the gold 
 bead. The solution of mercury oxide is easily obtained 
 by boiling the oxide with a strong solution of the cyanide 
 and filtering. According to Kuhn, the cyanide formed in 
 this manner has the formula HgCy-f 3HgO. The hydrated 
 nickel oxide precipitated may be filtered off, washed, dried, 
 ignited, and weighed. The cobalt is more readily and 
 conveniently estimated by difference, when, as it is always 
 possible, the two metals have been weighed together as 
 sulphates. We are not prepared to say that this modifi- 
 cation of Liebig's method of separating nickel and cobalt 
 gives better results than Stromeyer's process by means of 
 potassium nitrite, but it is at least very much more con- 
 venient, and requires much less time. The complete pre- 
 cipitation of cobalt in the form of Co 2 3 ,2N0 3 + 3KO,N03 
 usually requires at least forty-eight hours, and rarely 
 succeeds perfectly except in experienced hands. 
 M. Terreil has proposed a very excellent method for 
 separating these two metals. The author's method is 
 founded (1) on the insolubility of roseocobaltic hydro- 
 chlorate in acid liquids and ammoniacal salts, discovered 
 by M. Fremy ; (2) on the rapid transformation of ordinary 
 salts of cobalt into roseocobaltic salts, under the double 
 
 * * Chemical News,' March 17, 1865. 
 
848 THE ASSAY OF NICKEL AND COBALT. 
 
 influence of ammonia and oxidising bodies such as potas- 
 sium permanganate and alkaline hypoehlorites ; (3) on 
 the complete precipitation of manganese in ammoniacal 
 liquids by alkaline hypochlorites, and potassium perman- 
 ganate. 
 
 To separate cobalt from nickel, operate in the follow- 
 ing manner : 
 
 To the solution of the two metals add an excess of 
 ammonia, which re-dissolves the two oxides ; add to the 
 hot ammoniacal liquid a solution of potassium permanga- 
 nate, sufficient to cause the liquid to remain coloured 
 violet for a few instants by the excess of permanganate. 
 Boil the liquid for a few minutes, then add a slight excess 
 of hydrochloric acid, to re-dissolve the manganese oxide 
 which will have formed. Heat the liquid gently for twenty 
 or twenty-five minutes ; then let it stand for about twenty- 
 four hours. All the cobalt will then be deposited in the 
 form of a beautiful red-violet crystalline powder ; the pre- 
 cipitate is roseocobaltic hydrochlorate, which collect on a 
 weighed filter ; wash it on the filter with cold water, then 
 with diluted hydrochloric acid, or with a solution of am- 
 moniacal salt, and then with ordinary alcohol, which frees 
 it from ammoniacal salt. Dry it at 110, and weigh. 100 
 parts of roseocobaltic hydrochlorate correspond to 22*761 
 of metallic cobalt, or to 28*959 of protoxide of cobalt. 
 
 It is, however, better to take a given quantity of the 
 roseocobaltic salt, and reduce it by dry hydrogen ; this 
 leaves perfectly pure cobalt to be weighed. 
 
 Next boil the solution containing nickel to expel the 
 alcohol which has been introduced in washing the cobaltic 
 salt ; saturate it with ammonia, add another small excess 
 of permanganate of potash, and boil. All the manganese 
 will be precipitated ; filter the liquid, and all the nickel 
 will be found in the filtrate, from which it may easily be 
 separated in the state of sulphide, and then transformed 
 into oxide. 
 
 By this process the presence of a ten-thousandth part 
 of cobalt in a salt of nickel may be ascertained. 
 
 In this operation an alkaline hypochlorite may take 
 
DETECTION OF NICKEL BEFORE THE BLOWPIPE. 849 
 
 the place of the potassium permanganate, but then the 
 deposit of roseocobaltic salt takes place with extreme 
 slowness, and several days are required to complete it. 
 This reagent is preferable to permanganate when manga- 
 nese is to be separated from nickel and cobalt. 
 
 Quantitative Assay of Small Proportions of Cobalt in 
 Nickel. The following method is proposed by Dr. Fleit- 
 mann. It is well known that from solutions containing 
 both these metals the cobalt is first precipitated by hypo- 
 chlorites, as a brown hydrate, and the nickel hydrate does 
 not fall until after a further addition. The partial pre- 
 cipitation with sodium hypochlorite is so conducted that 
 at least two parts of nickel may be thrown down to one of 
 cobalt. The proportion may be judged by the colour of 
 the solution of the precipitate. If this solution is decidedly 
 red, more of the precipitant must be added. After a 
 slight washing the precipitate is dissolved on the filter 
 with warm hydrochloric acid,. the excess of chlorine re- 
 moved by boiling, and the mixture of cobaltous and 
 nickelous oxides is precipitated from the warm solution 
 with potash lye. The precipitate is filtered, slightly washed 
 with water, dissolved in acetic or nitric acid, and the 
 cobalt is then precipitated in the ordinary manner with 
 potassium nitrite. 
 
 Detection of Nickel before the Blowpipe.- The following 
 is Plattner's method for detecting nickel when contained 
 in large quantities of cobalt : 
 
 Fuse in the oxidising flame a moderate quantity of 
 borax to a bead in the loop of platinum wire, with suffi- 
 cient oxide of cobalt to give an opaque glass ; remove the 
 assay, and prepare one or two similar beads, and place the 
 whole in a charcoal cavity, with a button of pure gold 
 weighing from fifty to eighty milligrammes. The operator 
 must now heat in the reducing flame until he is satisfied 
 that the whole of the nickel is in a metallic state ; the 
 charcoal during the action must be inclined alternately 
 backwards and forwards, so that the gold button may 
 flow through the melted glass, and form an alloy with the 
 reduced particles of nickel. When the golden globule 
 
 3 i 
 
850 THE ASSAY OF NICKEL AXD COBALT. 
 
 solidifies, it must be extracted with a forceps, placed be- 
 tween paper, and struck with a hammer, so as to detach 
 all the adhering vitreous parts. The auriferous button, 
 which has become more or less grey from the presence of 
 nickel, and also more brittle than pure gold, is now to be 
 mixed with microcosmic salt, and heated for some time in 
 the oxidising flame. If the borax-glass has not been in 
 the first instance oversaturated with oxide of cobalt, a 
 bead will now be obtained, which is coloured only by 
 oxide of nickel, and will therefore appear brownish-red 
 while hot, and when cold reddish yellow. Should por- 
 tions of oxide of cobalt be also reduced, as the cobalt is 
 oxidised before the nickel, either a blue glass, coloured 
 by oxide of cobalt, or a green one if some nickel was 
 also oxidised will be obtained. In either case the glass 
 must be separated from the button, mixed with more 
 microcosmic salt, and heated in the oxidising flame 
 until it acquires a tinge. If the borax-glass had not 
 been over-saturated at the commencement, the colour 
 now obtained will proceed from nickel, although the 
 cobalt oxide contains a trace only ; but if nickel oxide 
 be not present, the microcosmic bead remains perfectly 
 colourless. 
 
 In the assay of substances containing cobalt, nickel, 
 and zinc, Alex. Olassen (' Zeitschrift fur Anal. Chemie,' 
 1879, p. 189) proceeds as follows: 
 
 To the solution, rendered as neutral as possible, he adds 
 so much neutral potassium oxalate (1 part of the salt and 
 3 parts water) that the precipitate is redissolved. He then 
 adds, with stirring, acetic acid about equal to the volume 
 of the liquid. The precipitate becomes crystalline at a 
 temperature of 50 to 60 C., and the liquid remains clear. 
 The precipitate is washed with a mixture of equal volumes 
 of concentrated acetic acid, alcohol, and water. The dry 
 precipitate, after the filter has been burnt on a platinum 
 wire, is first ignited very slightly in a covered platinum 
 crucible, so that no particles may be expelled by the 
 escaping carbonic oxide, and finally are ignited in an 
 open crucible. 
 
NICKEL AND COBALT GLANCE. 851 
 
 Ores containing Sulphur, Arsenic, Nickel Cobalt, and 
 Iron. (Arsenical nickel glance, cobalt-glance, red and 
 white nickel pyrites, cobalt speiss, commercial nickel.) 
 
 In nickel and cobalt glance sulphur is present in great 
 quantity, and is rarely absent in the remaining ores. Its 
 estimation is best effected in a separate portion. For 
 the separation and estimation of the metals the finely 
 powdered sample is oxidised either with aqua regia or 
 with hydrochloric acid and potassium chlorate, and the 
 arsenic is estimated as directed for the arsenides and sulph- 
 arsenides of iron. 
 
 The liquid filtered from the arsenic sulphide is freed 
 from sulphuretted hydrogen by heat, the iron is oxidised 
 with potassium chlorate, the free chlorine expelled by 
 heat and the addition of a little alcohol, the liquid is 
 largely diluted, placed in a basin, and mixed with sodium 
 carbonate till the acid reaction becomes very faint. 
 Barium carbonate levigated in water to a fine paste is 
 then added till it lies at the bottom of the vessel. After 
 repeated stirring the iron is all precipitated. It is filtered 
 off, dissolved in dilute hydrochloric acid, sulphuric acid 
 is added, the precipitate of barium sulphate removed, 
 and the ferric oxide precipitated with ammonia. The 
 liquid containing the nickel and cobalt is freed from 
 soluble barium compounds by means of sulphuric acid, 
 and the barium sulphate is removed by filtration. 
 
 The filtrate, containing the nickel and cobalt, is placed 
 in a basin, supersaturated with potash, heated to a boil, 
 filtered, washed with hot water, and the mixture of both 
 oxides dissolved in acetic acid. To this liquid is added a 
 concentrated solution of potassium nitrite (if it contains 
 free potash an excess of acetic acid must be present), and 
 the whole is let stand for 24 hours. 
 
 A yellow precipitate of potassium cobaltic nitrite is 
 produced, which is filtered off and washed in the cold with 
 a saturated solution of potassium chloride. More potas- 
 sium nitrite is added to the filtrate to ascertain whether 
 anything further is deposited. The yellow precipitate is 
 digested with hydrochloric acid, in which it dissolves, 
 
 3 i 2 
 
852 THE ASSAY OF NICKEL AND COBALT. 
 
 diluted, filtered into a basin, and the cobalt oxide is pre- 
 cipitated by boiling with an excess of potash. The pre- 
 cipitate is washed hot, and when dry it is placed in a 
 porcelain crucible and strongly ignited in a current of 
 hydrogen. After being allowed to cool in the same current 
 it is weighed in the covered crucible as metallic cobalt. It 
 is then well washed with water, ignited again in the stream 
 of hydrogen, and finally weighed. 
 
 In the liquid filtered from the cobaltic precipitate the 
 nickel is thrown down by boiling with potash. When care- 
 fully washed, dried, and ignited, it yields pure nickel oxide, 
 from the weight of which that of the metal is calculated. 
 
 Many substances of this class contain other constituents 
 in smaller proportion. 
 
 Commercial nickel contains silicon, which on dissolving 
 the metal in nitric acid, &c., is separated out as silica. For 
 its estimation the whole is evaporated to dryness in the 
 water-bath, the residue when cold is moistened with acid 
 and treated with water. The silica is then filtered off, and 
 the other constituents are estimated in the filtrate in the 
 usual manner. 
 
 The ores of nickel and cobalt and speiss-nickel contain 
 antimony. It is precipitated by sulphuretted hydrogen 
 along with arsenic. The precipitate is dissolved in aqua 
 regia, and the two metals are separated in the usual 
 manner. 
 
 The above ores and furnace products may also contain 
 copper, bismuth, and lead. In presence of small quantities 
 of these metals sulphuretted hydrogen should be passed 
 through the solution of the mixture for a rather shorter 
 time at first. These are precipitated before arsenic, but are 
 accompanied by a part of it. The precipitate is filtered off, 
 and the passage of the sulphuretted hydrogen through the 
 filtrate is continued till the arsenic is completely deposited. 
 
 The precipitated sulphides whilst still moist are placed 
 in a flask together with the filter, and are digested for a 
 considerable time with concentrated yellow ammonium 
 hydrosulphide. When cold the solution is diluted and 
 filtered with exclusion of air, washed with water, to which 
 
ALLOYS OF COPPER, ZINC, AND NICKEL. 853 
 
 a few drops of ammonium hydrosulphide have been added ; 
 the filtrate is slightly supersaturated with hydrochloric 
 acid, and the precipitate of arsenic sulphide is filtered off 
 and added to the main arsenical precipitate. The sulphides 
 (copper, bismuth, and lead) if their quantity permits are 
 separated by methods already given. If the quantities of 
 these metals are considerable, as is the case in many furnace 
 products, their assay is conducted as pointed out for lead- 
 and copper-speiss. Rammelsberg . 
 
 Alloys of Copper, Zinc, and Nickel The assay is con- 
 ducted as has been directed for alloys of copper and zinc. 
 The solution filtered from the copper sulphide, and con- 
 taining the zinc and nickel, is concentrated by evaporation 
 in order to remove the excess of sulphuretted hydrogen. 
 It is then poured into a flask, supersaturated with pure 
 potash and hydrocyanic acid enough to dissolve the whole 
 to a yellow liquid. From this solution the zinc is precipi- 
 tated as zinc sulphide by means of potassium monosulphide 
 (prepared by reducing potassium sulphate with charcoal), 
 and caused to settle by digestion. It is then passed 
 through a covered filter, the filtrate being collected in a 
 flask, washed with cold water, to which a little potassium 
 sulphide has been added, redissolved by digestion with 
 hydrochloric acid in a covered beaker till the odour 
 of sulphuretted hydrogen has entirely disappeared ; the 
 diluted solution is filtered into a capsule, and the zinc 
 oxide is thrown down by means of sodium carbonate. 
 
 The filtrate from the zinc sulphide is boiled in the flask 
 with aqua regia till the odours both of sulphuretted 
 hydrogen and hydrochloric acid are expelled, and do not 
 return on the addition of a little acid. The liquid is then 
 placed in a basin, supersaturated with potash, and kept at 
 a boil for a few minutes. The hydrated nickel-oxide is 
 washed with hot water, dried, ignited, and weighed, the 
 proportion of the metal being calculated from the weight 
 of the nickel-oxide. 
 
 In the electrolytic assay of copper and nickel Herpin 
 proceeds as follows in the assay of alloys of these metals. 
 He dissolves 1 grm. of the sample in nitric acid in a flask 
 
854 THE ASSAY OF NICKEL AND COBALT. 
 
 capable of holding 250 c.c., evaporates almost to dry ness, 
 and adds 4-5 c.c. sulphuric acid, and water enough to 
 make up a volume of 60-70 c.c. The liquid is rinsed into 
 a platinum capsule and submitted to electrolysis. The 
 copper only is deposited from an acid solution. 
 
 The liquid, still containing the nickel, is poured into a 
 flask like the one used for dissolving the alloy ; the plati- 
 num capsule is rinsed first with water, then with alcohol, 
 dried and weighed to estimate the copper. 
 
 The nickeliferous liquidate the washings, is heated to 
 a boil, partially neutralised with sodium carbonate, and 
 supersaturated with ammonia till it takes a blue colour. 
 It is then placed in the platinum capsule, and submitted to 
 electrolysis. Traces of lead and iron do not interfere. 
 
 W. Ohl (' Zeitschrift fur Anal. Chemie,' 1879, 523) gives 
 the following process for the assay of a nickel speiss : 
 1 grm. finely ground is placed in a beaker holding 300 c.c., 
 and covered with nitric acid or aqua regia. The beaker 
 is covered with a watch-glass and set on the sand-bath. 
 When it is completely dissolved the watch-glass is taken 
 off and the liquid evaporated to dryness. About 5 c.c. of 
 pure concentrated hydrochloric acid are added, and after 
 the mass is dissolved the beaker is half-filled with water. 
 When the solution is hot, sulphuretted hydrogen is passed 
 through it till cold. It is again set to warm, and again 
 treated with the same gas till cold. The precipitate of 
 copper and arsenic is quickly deposited, and the super- 
 natant liquid becomes clear. As arsenic sulphide is slightly 
 soluble in water containing sulphuretted hydrogen, the 
 beaker is set in a warm place till the smell becomes very 
 faint. If the precipitate is a fine uniform yellow it is 
 washed on filtering with cold pure water ; if darker, and 
 therefore containing more water, it is washed with sulphu- 
 retted hydrogen water. 
 
 The filtrate containing cobalt and nickel is evaporated 
 to dryness in a capsule holding f litre, adding a little 
 potassium chlorate to oxidise iron. The residue is taken 
 up with a little hot water and hydrochloric acid, preci- 
 pitated with pure solution of soda till the reaction is 
 
LEAD- AND COPPEK-SPE1SS. 355 
 
 alkaline, redissolved in pure acetic acid, largely diluted 
 and heated to a boil. The iron is all deposited as basic 
 ferric acetate, which is filtered off and washed with hot 
 water till a drop of ammonium sulphide produces no 
 turbidity in a drop of the washings. The solution, freed 
 from the iron, is evaporated to dryness, dissolved in water 
 and a few c.c. of dilute sulphuric acid, placed in a beaker 
 holding 600 c.c., supersaturated with ammonia, and sub- 
 mitted to the electric current. 
 
 When the electrolysis is complete, a drop of the liquid 
 is withdrawn with a pipette, filtered, and mixed with a 
 drop of ammonium sulphide. If no turbidity is formed the 
 platinum cone is withdrawn, washed first in water and then 
 in absolute alcohol, and dried. The increase of weight 
 gives the sum of the cobalt and nickel. 
 
 LEAD- AND COPPEE-SPEISS. 
 
 These substances are very complicated products formed 
 during the metallurgical treatment of arseniferous and 
 antimoniferous ores of lead and copper. They may con- 
 tain copper, lead, iron, nickel, cobalt, zinc, bismuth, silver, 
 arsenic, antimony, and sulphur. 
 
 A. Speiss containing little or no Lead or Antimony 
 One portion is taken for the determination of sulphur, 
 and another is dissolved in aqua regia, or in a mixture of 
 hydrochloric acid and potassium chlorate. The proportion 
 of silver is generally so small that no silver chloride 
 remains. If any lead chloride is separated it is dissolved 
 by heating in water. The solution is treated with sulphu- 
 retted hydrogen, when lead, copper, bismuth, antimony, 
 and arsenic are deposited. To ascertain that the arsenic 
 is completely precipitated the liquid is heated and again 
 treated with sulphuretted hydrogen. The metallic sul- 
 phides are digested with concentrated yellow ammonium 
 hydrosulphide, in order to dissolve the arsenic and anti- 
 mony sulphides. 
 
 After filtration both are precipitated by hydrochloric 
 acid, and separated as directed under arsenical iron. 
 
856 THE ASSAY OF NICKEL AND COBALT. 
 
 The undissolved sulphides of copper, lead, and bismuth 
 are allowed to become air-dry, and are then detached from 
 the filter, which is incinerated in a small porcelain crucible. 
 First, these ashes, and then the total sulphides, are dis- 
 solved in nitric acid. The sulphur, which is not quite 
 pure, is filtered off and gently heated in a porcelain 
 crucible, the slight residue being digested again with nitric 
 acid and added to the main solution (if lead sulphate 
 remains unattacked it must be collected on a weighed 
 filter). The nitric solution of the three metals is concen- 
 trated by evaporation, and the lead separated as already 
 directed. The filtrate is saturated with sodium carbonate, 
 potash is added, the whole heated to a boil, the oxides 
 are filtered off, washed slightly, dissolved in the smallest 
 quantity of hydrochloric acid, and the bismuth is precipi- 
 tated by the addition of much water as basic oxychloride, 
 which is reduced by potassium-cyanide, as already de- 
 scribed. Copper is precipitated from the filtrate by sul- 
 phuretted hydrogen. 
 
 In the filtrate from the first sulphuretted hydrogen 
 precipitate there are still found iron (as ferrous oxide), 
 nickel, cobalt, and zinc. The liquid is concentrated to 
 expel sulphuretted hydrogen and most of the free acid ; 
 the iron is peroxidised by the addition of a little potassium 
 chlorate, diluted, the free chlorine expelled by heating 
 with a few drops of alcohol, and the iron oxide is separated 
 from zinc, nickel, and cobalt as directed for arsenical 
 nickel glance, &c. 
 
 B. Speiss containing much Lead or Antimony. The 
 process to be followed is tedious, and requires much care. 
 The substance, finely powdered, is decomposed by heating 
 in a current of chlorine gas. 
 
 The apparatus requisite consists of a capacious flask, in 
 which the necessary chlorine gas is evolved from a mixture 
 of 2 parts black oxide of manganese, 3 parts of common 
 salt and dilute sulphuric acid (1J to 2 parts water to 1 
 part of the monohydrated acid). The flask is closed with 
 a cork, through which passes a tube bent twice at right 
 angles. Its longer leg dips into concentrated sulphuric 
 
LEAD- AND COPPEK-SPEISS. 857 
 
 acid contained in a second flask, where the gas is partially 
 dried, and whence it issues to pass by another tube into a 
 chloride of calcium apparatus, and thence into the bulb- 
 tube in which the reaction is to take place. The tube is 
 tared, from 1 to 2 grms. of the finely ground sample is 
 inserted into the bulb through its wider tube, both ends 
 are cleaned with a feather, so that none remains except in 
 the bulb, and the whole is weighed to find the exact quantity 
 which has been taken for analysis. The shorter tube is 
 then connected to the chloride of calcium apparatus, and 
 the longer, or bent tube, is conducted through a cork into 
 a receiver, which resembles a Liebig's bulb-tube on a larger 
 scale, and with parallel limbs. It is filled with dilute 
 hydrochloric acid, to which a little tartaric acid is added, 
 if the body under analysis contains antimony. From the 
 second limb of this receiver a bent tube passes into a 
 Woolf's bottle, and dips into the same solution. From 
 the other aperture passes a tube which carries away the 
 escaping chlorine. When all parts of the apparatus are 
 full of chlorine heat is carefully and gradually applied to 
 the bulb, which must not reach visible redness. After the 
 completion of the operation the bulb contains the non- 
 volatile lead, silver, bismuth, copper, nickel, cobalt chlo- 
 rides, and in part those of iron and zinc. They are dis- 
 solved out by means of water and hydrochloric acid ; 
 undecomposed portions of the sample and silver chloride 
 may remain, and are collected upon a weighed filter, and 
 after weighing treated with ammonia, which dissolves the 
 silver chloride, leaving the undecomposed matter behind. 
 The hydrochloric solution is filtered into dilute sulphuric 
 acid, evaporated, and the lead separated as lead sulphate. 
 The filtrate is precipitated with sulphuretted hydrogen, 
 and copper and bismuth are separated by dissolving the 
 sulphides in hydrochloric acid 5 to which a little nitric acid 
 is added, the solution concentrated to a small bulk, the 
 bismuth precipitated by water, the basic bismuth chloride 
 reduced by potassium cyanide, and in the filtrate the 
 copper precipitated by sulphuretted hydrogen in the ordi- 
 nary manner. The filtrate from which the copper and 
 
858 THE ASSAY OF NICKEL AND COBALT. 
 
 bismuth sulphides were removed contains iron, zinc, nickel, 
 and cobalt, and is treated as already pointed out. 
 
 From the volatile chlorides contained in the liquid 
 from the receivers the sulphuric acid is removed by barium 
 chloride, and any excess of the latter by the cautious addi- 
 tion of sulphuric acid. Antimony and arsenic are precipi- 
 tated by sulphuretted hydrogen, and separated in the usual 
 manner. The filtrate contains the volatilised portions of 
 iron and zinc, which are also dealt with by ordinary 
 methods. 
 
CHAPTEE XXIV. 
 
 THE ASSAY OF SULPHUE. 
 
 THE only commercially valuable Sulphur-minerals are : 
 
 I. Sulphurous Earth (native sulphur). 
 
 In Sicily these minerals are divided into five classes : 
 
 1. Very rich ores, containing 32 34 per cent, sulphur 
 
 2. Kich 2426 
 
 3. Good 16-18 
 
 4. Middling ,, 8 9 
 
 5. Poor 3 5 
 
 II. Iron and Copper Pyrites (FeS 2 ), and (Cu 2 S, Fe 2 S 3 ). 
 
 These ores are used to a very large extent for the 
 manufacture of sulphuric acid. 
 
 In order to approximately estimate the value of ores 
 of the first class for such manufacture, the following method 
 of assay may be used. 
 
 Assay by Distillation. A certain quantity of the pul- 
 verised sulphurous earth is heated in a glass retort, which 
 is furnished with a receiver. The retort is then heated, 
 gradually raising the temperature, till no more sulphur is 
 evolved. The latter will collect in all cases in the neck of 
 the retort and receiver, which may be of glass and must 
 be kept cool. 
 
 The sulphur derived from sulphurous earth is generally 
 pure, whilst that from pyrites frequently contains arsenic 
 and selenium, and sometimes traces of thallium. 
 
 ASSAY OF IRON AND COPPER PYRITES. 
 
 A. Assay of Sulphur in the Dry Way. Fuse the 
 weighed ore with a weighed quantity of anhydrous sodium 
 carbonate, twice as much potassium chlorate as ore, and 
 
860 THE ASSAY OF SULPHUK. 
 
 from 12 to 20 times as much sodium chloride (added to 
 moderate the action) ; carbonic acid is expelled, potassium 
 chloride formed, and all the sulphur converted into sodium 
 sulphate. By dissolving the residue in water and esti- 
 mating alkalimetrically the unaltered sodium carbonate by 
 a standard acid solution, the portion converted into sul- 
 phate, and hence the sulphur in the ore, is known. Besides 
 the difficulty of preventing loss by deflagration, this method 
 is open to the small errors caused by the reckoning all 
 arsenic present to be sulphur : this, however, is usually of 
 no moment for commercial purposes ; and calcium carbo- 
 nate in the ore may, if required, be previously dissolved 
 out by dilute hydrochloric acid. 
 
 In performing fusions of sulphur compounds with 
 nitre or potassium chlorate the operator must bear in mind 
 a source of error, first pointed out by Dr. David S. Price, 
 in consequence of sulphur compounds being contained in 
 the coal-gas which frequently serves as fuel in these experi- 
 ments. By exposing a small quantity of fused nitre, on 
 the outside of a platinum capsule, to the flame of a Bunsen 
 gas-burner for three-quarters of an hour, Dr. Price suc- 
 ceeded in detecting the presence of sulphuric acid to an 
 amount equivalent to 12 milligrammes of sulphur. This 
 sulphuric acid had been formed by the oxidation of the 
 sulphur in the coal-gas, and when dissolved in water it 
 gave an immediate precipitate with chloride of barium. 
 By making a similar experiment with the use of a spirit- 
 lamp as the source of heat, no trace of potassium sulphate 
 was formed ; nor was any appreciable amount of sulphuric 
 acid generated in another trial made by fusing a small 
 quantity of nitre inside a platinum capsule heated over 
 gas ; but whenever the fused salt crept over the edges of 
 the capsule, some potassium sulphate was sure to be formed. 
 This observation may become a matter of importance when 
 the amount of sulphur in pig-iron is estimated by fusion 
 with pure nitre, for the author has remarked that samples 
 containing much manganese are especially liable to impart 
 to the fused salt a tendency to creep up and escape over the 
 sides of the crucible. 
 
ASSAY OF SULPHUR IN PYRITES. 861 
 
 B. Assay of Sulphur in the Wet Way. Mr. C. E. A. 
 Wright recommends the following process as being the one 
 best adapted for commercial purposes : A known weight 
 of the ore reduced to fine powder is oxidised (best in a 
 small flask with a funnel in the mouth to avoid loss by 
 spirting, and heated on a sand-bath), either by strong 
 nitric acid, or aqua regia, perfectly free from sulphuric 
 acid ; after the oxidation is complete, the liquid is evapo- 
 rated down as far as possible to expel the majority of the 
 remaining nitric or hydrochloric acid ; the residue is boiled 
 with a little water, and almost but not quite neutralised by 
 ammonia ; a solution of barium chloride of known strength 
 is then added until no further precipitate is produced, the 
 exact point being found by filtering off a little of the liquid 
 after each addition of barium chloride, and adding to it a 
 few more drops of the standard solution, care being always 
 taken, in case of a further precipitate being thus produced, 
 to add this filtrate to the original solution, and mix well 
 before filtering a second time. In case of overstepping the 
 mark, it is convenient to have at hand a solution of sodium 
 sulphate of strength precisely equal to that of the barium 
 chloride ; this solution may then be cautiously added 
 with repeated filtration and examination of the filtrate 
 with the sulphate solution, until the point is just reached 
 when addition of sulphate solution produces no further 
 precipitate ; by subtracting the volume of sulphate solu- 
 tion thus used from the total volume of barium solution 
 added, the exact quantity of this latter consumed is 
 known. If 1 grm. of sulphur ore be taken, and 32 -5 
 grms. of pure anhydrous barium chloride be dissolved in 
 a litre of fluid, each cubic centimetre of barium solution 
 used will represent 0*5 per cent, of sulphur in the ore 
 examined : 22-19 grms. of anhydrous sodium sulphate 
 being dissolved to a litre for the second solution. In case 
 of lead being contained in the ore, an error is introduced 
 from the formation of insoluble lead sulphate. As lead, 
 however, rarely occurs in any perceptible quantity, this 
 error is negligible, the process only giving approximate 
 results. 
 
86*2 THE ASSAY OF SULPHUR. 
 
 Where greater accuracy is required, it is advisable to 
 precipitate the sulphuric acid formed from the original 
 liquid (filtered from insoluble residue) by barium nitrate 
 or chloride, and to weigh the barium sulphate produced. 
 Instead of oxidising by acids, the powdered ore may be 
 suspended in caustic potash (free from sulphate), and 
 oxidised by passing washed chlorine into the liquid ; lead, 
 being converted into dioxide, is thus rendered non-injurious ; 
 the alkaline liquid obtained is acidified, and precipitated 
 by chloride of barium as before. In the volumetric es- 
 timation usually pursued, a curious circumstance is occa- 
 sionally observable when much free acid exists in the 
 solution viz. that a point may be reached when the fil- 
 tered liquid is clear, and remains so even on standing for 
 a short time, but yields a cloud, or even a precipitate, on 
 the addition either of barium solution or sulphate solution. 
 This source of error is mostly avoidable by nearly neutral- 
 ising the free acid with ammonia. 
 
 Instead of chlorine, hypochlorous acid may be used to 
 transform the sulphur of pyrites into sulphuric acid, which 
 is then estimated by barium chloride. Finely pulverise 
 the mineral and suspend it in water, through which a 
 current of gaseous hypochlorous acid, or, better still, 
 hypochloric acid, is passed ; this entirely dissolves the 
 pyrites. Hypochlorous acid is prepared by heating a milk 
 of calcium carbonate through which a current of chlorine 
 is passed to saturation. Hypochloric acid is obtained by 
 heating in a water-bath a tube, supplied with a cork and 
 delivery tube, and containing a mixture of nine equivalents 
 of oxalic acid and one equivalent of chlorate of potash. 
 
 Mr. A. H. Pearson has given the following very accu- 
 rate method of estimating sulphur in pyrites : Weigh out 
 1 grm. or less of the powdered ore ; place the powder in a 
 porcelain dish, together with a small quantity of potassium 
 chlorate ; pour upon it some 50 c.c. of pure nitric acid of 
 39 B., and cover the mixture with an inverted glass funnel 
 with bent stem. Set the dish upon a water-bath and 
 heat the water to boiling. From time to time throw 
 crystals of potassium chlorate into the hot acid. By adding 
 
ASSAY OF SULPHUR IN PYRITES. 863 
 
 rather large crystals of the chlorate at frequent intervals, 
 it is easy to oxidise the whole of the sulphide in half an 
 hour ; but since the solution obtained in that case is 
 highly charged with saline matter, it will usually be found 
 more advantageous to use less of the potassium chlorate, 
 and to allow a somewhat longer time for the process of 
 oxidation. 
 
 When all the sulphur has been oxidised, rinse the 
 funnel with water and remove it from the dish. Evapo- 
 rate the liquid to a small bulk, then add to it a little 
 concentrated hydrochloric acid, and again evaporate to 
 absolute dryness, in order to render silicic acid insoluble. 
 Moisten the residue with concentrated hydrochloric acid, 
 mix it with water, and filter to separate silicic acid and 
 gangue. 
 
 To the filtrate from the silicic acid add a quantity of 
 solid tartaric acid, about as much as that of the pyrites 
 originally taken ; heat the liquid almost to boiling, and 
 add to it an excess of barium chloride, to precipitate the 
 sulphuric acid. After the barium sulphate has been 
 allowed to subside, wash it thoroughly by decantation, 
 first with hot water, and afterwards with a dilute solution 
 of ammonium acetate (the latter may be prepared at the 
 moment of use by mixing ammonia-water and acetic acid). 
 The purpose of the ammonium acetate is to dissolve any 
 barium nitrate which may adhere to the sulphate : that of 
 the tartaric acid is to prevent the precipitation of iron 
 compounds together with the barium sulphate. In an 
 experiment where 0'7 grm. of pyrites was oxidised with 
 potassium chlorate and nitric acid, and the filtrate from 
 silica was acidulated with hydrochloric acid without the 
 addition of tartaric acid, there was thrown down, on the 
 addition of barium chloride, a bright yellow precipitate, 
 which became darker-coloured when the solution was 
 boiled. It was not only found to be impossible to wash 
 out the iron with which this precipitate was contaminated, 
 but the consistency of the precipitate was such that it was 
 a difficult matter even to wash away the saline liquor in 
 which it was formed. 
 
864 THE ASSAY OF SULPHUK. 
 
 M. A. Houzeau, in his so-called gravi-volumetric 
 process, attacks 1 gramme of the powdered ore with a 
 mixture of 4 parts pure potassium nitrate, and 3 parts 
 sodium carbonate, likewise pure. The saline mass is 
 dissolved in hot water, and the ferric oxide filtered from 
 the alkaline sulphate. The washing-waters are added to 
 the filtrate, and, after cooling, it is made up with distilled 
 water to half a litre. A portion is then taken (10 c.c.) 
 and acidulated with a few drops of pure acetic acid, and 
 the sulphuric acid is rapidly estimated, as the author 
 has elsewhere indicated for selenitic waters, by making use 
 of a standard solution of barium chloride, applied by the 
 aid of the gravi-volumeter in place of the ordinary burette,, 
 the use of which, in such estimations, yields merely erro- 
 neous results. In the gravi-volumeter the weight of the 
 standard solution shows the quantity of the reagent which 
 has been used. But each drop of the barytic liquid 
 deli vered by the gravi-volumeter weighs at the tempera- 
 ture of 13 exactly 0-050 grm. 
 
 In the Frieberg works, 1 grm. of finely ground ore is 
 mixed with 3 grms. anhydrous sodium carbonate, and an 
 equal weight of saltpetre. This mixture is placed in an 
 iron crucible, melted in a muffle. At a red heat, the mass 
 is dissolved in hot water, and the liquid is filtered into 
 a beaker, in which there is a little hydrochloric acid to 
 saturate the excess of soda. The liquid, which should 
 have an acid reaction, is then boiled for a short time,, 
 and the sulphuric acid is estimated volumetrically with a 
 solution of barium chloride, standardised so that 1 c.c. 
 indicates 2 per cent, of sulphur. 
 
 B. Deutecom adopts the following process : 1 grm. 
 pyrites is mixed in a large covered crucible with 8 grms. 
 of a mixture of equal parts potassium chlorate, sodium 
 carbonate, and sodium chloride. The crucible is heated at 
 first gently, so as to dry the contents, which are after- 
 wards melted at a high temperature. The mass, when 
 cold, is treated with boiling water, and the solution, to- 
 gether with the deposit, is introduced into a measuring- 
 flask of 200 c.c., filled up, filtered, and the sulphuric acid 
 
ASSAY OF SULPHUR IN PYRITES. 865 
 
 is estimated in aliquot parts, say 50 c.c. The insoluble 
 residue does not retain any sulphuric acid. In this manner 
 the use of nitric acid is avoided. The decomposition of 
 the potassium chlorate is complete. 
 
 In assaying pyrites for sulphur only by fusion, Mr. 
 P. Holland has obtained good results by the following 
 process, which may be useful in such laboratories as do 
 not possess large platinum crucibles. A test-tube, or piece 
 of sealed combustion tube, about 6 inches long and 
 half an inch in internal width, is fitted with a cork and 
 delivery tube, the latter bent at a right angle, and long 
 enough to reach to the bottom of the flask in which it is 
 intended to make the titration. The fusion mixture con- 
 sists of equal parts of nitre and ignited sodium bicar- 
 bonate, both free from sulphur, dry, and in fine powder. 
 Nine to ten grammes are taken in an operation, together 
 with one of pyrites ; the latter must be in exceedingly fine 
 powder. The two are mixed in a warm porcelain dish or 
 agate mortar, and transferred to the tube without loss. 
 The delivery tube is then inserted, with its extremity 
 dipping into the flask. A channel is made on the surface 
 of the mixture, and the tube, suitably supported, is heated 
 by small portions at a time with a Bunsen gas flame, com- 
 mencing as usual with the anterior portion. When the 
 operation is progressing favourably, the deflagration 
 proceeds for a few seconds after removing the flame. 
 
 There is no danger to be apprehended, and the tube 
 does not crack or blow out with proper care. When the 
 tube has been heated throughout, and the deflagration 
 has ceased, it is then more strongly heated with a Hera- 
 path or powerful gas flame. It is a good plan at this 
 stage to slip a coil of wire gauze over the tube, which 
 helps to accumulate the heat. It is not, however, neces- 
 sary that the contents should be fused a second time ; at 
 least this has not been done in the experiments appended. 
 The sulphur ores examined have yielded their sulphur 
 readily. 
 
 The gaseous products of the combustion, which 
 mechanically carry over with them small quantities of 
 
866 THE ASSAY OF SULPHUR. 
 
 sulphates or sulphuric acid, \being heavier than air, collect 
 in the flask, and are washed by shaking with a little 
 water, closing the flask with the palm of the hand. The 
 delivery tube is also washed. That containing the fused 
 mass is carefully broken and put in the flask, together 
 with sufficient hydrochloric acid to dissolve nearly the 
 whole of the iron oxide ; then ammonia is added, until a 
 precipitate of oxide reappears ; and lastly, as much hydro- 
 chloric acid and water as is necessary to bring the fluid 
 to the conditions which were obtained when the barium 
 solution was standardised. The author has used 2 c.c. of 
 free acid, and the total volume of solution was 200 c.c. 
 
 F. Bceckmann recommends the following modification 
 of the potassium chlorate method : Half a grm. of finely 
 ground pyrites (sifting is not absolutely necessary) is 
 mixed in a large platinum capsule with the well-known 
 mixture of 6 parts sodium carbonate and 1 part potassium 
 chlorate. The mixing is effected with a platinum spatula, 
 and is then made more complete by gentle rubbing with 
 an agate pestle fixed to a wooden handle. The whole is 
 then fused over the blast-lamp. The aqueous solution of 
 the melt is first poured into a beaker to avoid spirting, 
 and thence into another tall beaker containing an excess of 
 hydrochloric acid. The filtered solution is heated and 
 precipitated with hot barium chloride, heated gently upon 
 the sand-bath for a time until the liquid standing above 
 the precipitate has become clear, and is filtered at once. 
 The burnt ores in sulphuric acid works have been for a 
 long time assayed for sulphur by this process. Take 
 about 2 grms. of burnt ore to from 20 to 25 grms. of 
 chlorate mixture. 
 
867 
 
 CHAPTER XXV. 
 
 DISCRIMINATION OF GEMS AND PRECIOUS STONES. 
 
 SIMPLE characteristics of and means of recognising many 
 gems and precious stones have been given at page 291, in 
 the section on the discrimination of minerals. 
 
 The present chapter contains much information which 
 could not appropriately be introduced in the previous 
 chapter, which was intended chiefly for the use of tra- 
 vellers and explorers. Some trifling repetitions occur 
 purposely to save the inconvenience of referring back. 
 
 The principle sources of recognition are colour, crys- 
 talline form, specific gravity, and hardness. In the present 
 chapter will be introduced all the most constantly oc- 
 curring natural forms of the gems and precious stones 
 mentioned. 
 
 The specific gravity or density of a substance is the 
 proportion of its weight to its volume, and it forms a 
 characteristic property of substances. A full description 
 of the method of taking specific gravities has been given 
 at pages 241, 242, 243, 244. 
 
 COLOURLESS STONES. 
 
 THE DIAMOND. 
 
 (See also p. 257.) 
 
 Specific gravity, 3'48 to 3-52 ; hardness, 10. The 
 diamond is the hardest of all known substances. It is 
 the only substance which is capable of cutting glass, 
 although most gems will scratch glass ; hence it is the: 
 
 3 K 2 
 
868 GEMS AND PRECIOUS STONES. 
 
 utmost term of hardness. When cut and polished, it 
 is the most brilliant gem. It occasionally becomes phos- 
 phorescent on exposure to light. The greater number of 
 diamonds are limpid and colourless, but many coloured 
 specimens are found ; as rose, yellow, orange, blue, green, 
 brown, or even black. It occurs in regular crystals, 
 octahedrons, dodecahedrons, and more complex forms ; see 
 figs. 131, 132, 133, 134. 
 
 The crystalline faces are often curved. The cleavage 
 is octahedral and highly perfect ; hence, although dia- 
 monds are so exceedingly hard, they are somewhat brittle, 
 owing to their tendency to facile cleavage. Like most gems, 
 
 FIG. 144. FIG. 145. 
 
 FIG. 147. 
 FIG. 146. 
 
 they become electrical by friction ; but it has been re- 
 marked that other gems do not, unless they have been 
 previously polished. 
 
 Composition (C) : Pure carbon. 
 
 The Matrix of the Diamond. 
 
 Professor H. Carvill Lewis gives the following interest- 
 ing notes on the matrix of the diamond, arrived at from a 
 microscopical study of the remarkable porphyritic perido- 
 tite which contains the diamonds in South Africa. 
 
MATRIX OF THE DIAMOND. 869 
 
 The olivine, forming much the most abundant con- 
 stituent, is in porphyritic crystals, sometimes well bounded 
 by crystal faces, at other times rounded and with corrosive 
 cavities, such as occur in it in basaltic rocks. It rarely 
 encloses rounded grains of glassy bronzite, as has been 
 observed in meteorites. The olivine alters either into 
 serpentine in the ordinary way, or into an aggregate of 
 acicular tremolite crystals, the so-called 'pilit,' or becomes 
 surrounded by a zone of indigo blue bastite a new variety 
 of that substance. The olivine is distinguished by an un- 
 usually good cleavage in two directions. 
 
 Bronzite, Chrome diallage, and Smaragdite occur in 
 fine green plates, closely resembling one another. The 
 bronzite is often surrounded by a remarkable zone, with a 
 centric, pegmatitic, or chondritic structure, such as occurs 
 in certain meteorites. This zone is mainly composed of 
 wormlike olivine grains, but a mineral having the optical 
 characters of cyanite also occurs in this zone. 
 
 Biotite, a characteristic constituent, occurs in conspicu- 
 ous plates, often twinned, generally rounded, and distin- 
 guished by its weak pleochroism, a character peculiar to 
 the biotite of ultra-basic eruptive rocks. It alters by 
 decomposition into the so-called Vaalite. 
 
 Perofskite occurs in very numerous but small crystals, 
 which optically appear to be compound rhombic twins. 
 
 Pyrope is abundant in rounded red grains. Titanic 
 iron, chromic iron, and some fifteen other minerals are 
 also found. Eutile is formed as a secondary mineral 
 through the alteration of olivine into serpentine, being a 
 genesis of rutile not heretofore observed. 
 
 The chemical composition shows this to be one of the 
 most basic rocks known, and is a composition which by 
 calculation would belong to a rock composed of equal 
 parts of olivine and serpentine, impregnated by calcite. 
 
 The structure is at the same time porphyritic and brec- 
 ciated, being one characteristic of a volcanic rock which, 
 after becoming hard, had been subjected to mechanical 
 movements. It is a volcanic breccia, but not an ash or 
 tuff, the peculiar structure being apparently due to succes- 
 
870 GEMS AND PRECIOUS STONES. 
 
 sive paroxysmal eruptions. A similar structure is known 
 in meteorites, with which bodies this rock has several 
 analogies. A large amount of the adjoining bituminous 
 shale is enclosed, and has been more or less baked and 
 altered. The occurrence of minute tourmalines is evidence 
 of fumarole action. 
 
 The microscopical examination supports the geological 
 data in testifying to the igneous and eruptive character of 
 the peridotite, which lies in the neck or vent of an old 
 volcano. 
 
 While belonging to the family of peridotites, this rock 
 is quite distinct in structure and composition from any 
 member of that group heretofore named. It is more basic 
 than the picrite porphyrites, and is not holocrystalline 
 like dunite or saxonite. It is clearly a new rock type, 
 worthy of a distinctive name. The name Kimberlite, from 
 the famous locality where it was first observed, is there- 
 fore proposed. 
 
 Kimberlite probably occurs in several places in Europe, 
 certain garnetiferous serpentines belonging here. It is 
 already known at two places in the United States : at 
 Elliott County, Kentucky, and at Syracuse, New York ;at 
 both of which places it is eruptive and post-carboniferous, 
 similar in structure and composition to the Kimberley 
 rock. 
 
 At the diamond localities in other parts of the world 
 diamonds are found either in diluvial gravels or in con- 
 glomerates of secondary origin, and the original matrix 
 is difficult to discover. Thus, in India and Brazil the 
 diamonds lie in a conglomerate with other pebbles, and 
 their matrix has not been discovered. Recent observations 
 in Brazil have proved that it is a mistake to suppose that 
 diamonds occur in itacolumite, specimens supposed to show 
 this association being artificially manufactured. But at 
 other diamond localities, where the geology of the region 
 is better known than in India or Brazil, the matrix of the 
 diamond may be inferred with some degree of certainty. 
 Thus, in Borneo diamonds and platinum occur only in 
 those rivers which drain a serpentine district, and on Tanah 
 
MATRIX OF THE DIAMOND. 871 
 
 Laut they also lie on serpentine. In New South Wales, 
 near each locality where diamonds occur, serpentine also 
 occurs, and is sometimes in contact with carboniferous 
 shales. Platinum, also derived from eruptive serpentine, 
 occurs here with the diamonds. In the Urals, diamonds 
 have been reported from four widely separated localities, 
 and at each of these, as shown on Murchison's map, 
 serpentine occurs. At one of the localities the serpentine 
 has been shown to be an altered peridotite. A diamond 
 has been found in Bohemia in a sand containing pyropes, 
 -and these pyropes are now known to have been derived 
 from a serpentine altered from a peridotite. In North 
 Carolina a number of diamonds and some platinum have 
 been found in river sands, and that state is distinguished 
 from all others in eastern America by its great beds 
 of peridotite and its abundant serpentine. Finally, in 
 northern California, where diamonds occur plentifully 
 .and are associated with platinum, there are great out- 
 bursts of post-carboniferous eruptive serpentine, the ser- 
 pentine being more abundant than elsewhere in North 
 America. At all the localities mentioned chromic and 
 titanic iron ore occur in the diamond-bearing sand, and 
 both of these minerals are characteristic constituents of 
 serpentine. 
 
 All the facts thus far collected indicate serpentine, 
 in the form of a decomposed eruptive peridotite, as the 
 original matrix of the diamond. 
 
 Speaking of diamonds, Professor Orton says that few 
 things are so unpromising and unattractive as these gems 
 in their native state. Hence their slow discovery. There 
 is little doubt that diamonds exist in many places as yet 
 unknown, or where their presence is unsuspected. It is 
 very difficult for the unpractised eye to distinguish them 
 from crystals of quartz or topaz. The colour constitutes 
 the main difficulty in detecting their presence. They are 
 of various shades of yellowish -brown, green, blue, and rose- 
 Ted, and thus closely resemble the common gravel by which 
 they are surrounded. Often they are not unlike a lump 
 of gum arabic, neither brilliant nor transparent. The 
 
872 GEMS AND PRECIOUS STONES. 
 
 finest, however, are colourless, and appear like rock 
 crystals. 
 
 In Brazil, where great numbers of diamonds, chiefly of 
 small size, have been discovered, the method of searching 
 for them is to wash the sand of certain rivers in a manner 
 precisely similar to that employed in the gold fields 
 namely, by prospecting pans. A shovelful of earth is 
 thrown into the pan, which is then immersed in water, and 
 gently moved about. As the washing goes on, the pebbles r 
 dirt, and sand are removed, and the pan then contains 
 about a pint of thin mud. Great caution is now observed, 
 and ultimately there remains only a small quantity of sand. 
 The diamonds and particles of gold, if present, sink to the 
 bottom, being heavier, and are selected and removed by 
 the practised fingers of the operator. But how shall the 
 gems be detected by one who in a jeweller's shop could 
 not separate them from quartz or French paste ? The 
 difficulty can only be overcome by testing such stones as 
 may be suspected to be precious. Let these be tried by 
 the very sure operation of attempting to cut with their 
 sharp corners glass, crystal, or quartz. When too minute 
 to be held between the finger and thumb, the specimens 
 may be pressed into the end of a stick of hard wood and 
 run along the surface of window glass. A diamond will 
 make its mark, and cause, too, a ready fracture in the line 
 over which it has travelled. It will also easily scratch 
 rock crystal, as few crystals will. 
 
 But a more certain and peculiar characteristic of the 
 diamond lies in the form of its crystals. The ruby and 
 topaz will scratch quartz, but no mineral which will scratch 
 quartz has the curved edges of the diamond. In small crys- 
 tals this peculiarity can be seen only by means of a magnify- 
 ing glass ; but it is invariably present. Interrupted, convex,, 
 or rounded angles are sure indications of genuineness. 
 Quartz crystal is surrounded by six faces, the diamond 
 by four. The diamond breaks with difficulty ; and hence 
 a test sometimes used is to place the specimen between 
 two hard bodies, as a couple of coins, and force them 
 together with the hands. Such a pressure will crush a 
 
THE DIAMOND. 87$ 
 
 particle of quartz, but the diamond will only indent the 
 metal. 
 
 The imperfections of the diamond, and, in fact, of all 
 cut gems, are made visible by putting them into oil of 
 cassia, when the slighest flaw will be seen. 
 
 If a rough diamond resemble a drop of clear spring 
 water, in the middle of which you perceive a strong light ; 
 or if it has a rough coat, so that you can hardly see through 
 it, but white, and as if made rough by art, yet clear of 
 flaws or veins ; or if the coat be smooth and bright, with 
 a tincture of green in it, it is a good stone. If it has a 
 milky cast or a yellowish -green coat, beware of it. Eough 
 diamonds with a greenish crust are the most limpid when 
 cut. 
 
 Diamonds are found in loose pebbly earth, along with 
 gold, a little way below the surface, towards the lower 
 outlet of broad valleys, rather than upon the ridges of the 
 adjoining hills. 
 
 Prof. Silliman, on examining with the microscope a 
 small parcel of the sand resulting from the hydraulic treat- 
 ment of ores, found that they abounded in fine colourless 
 zircons, along with crystals of topaz, fragments of quartz,, 
 grains of chrome iron, and titanic acid, and globular bodies 
 of a very high refractive power, which he believes to be 
 diamonds. 
 
 Mr. J. Torry, in a single sample of the sands washed 
 from the gold ores of Nicaragua, found twenty mineral 
 species, some of them very rare. 
 
 For estimating the specific gravity of certain minerals,, 
 and separating diamond dust or small diamonds and other 
 gems from quartz, sand, &c., Mr. E. Sonstadt * uses a 
 solution in water of pure potassium iodide and pure mer- 
 curic iodide in cold water. It should be diluted to such 
 a strength that quartz will just float in it. (See ante> 
 pages 243, 244.) 
 
 * ' Chemical News,' March 20, 1874. 
 
874 
 
 GEMS AND PRECIOUS STONES. 
 
 QUARTZ. 
 
 Specific gravity, 2 -55 to 2*7 ; hardness, 7. Quartz 
 occurs in many forms, and has often by inexperienced 
 persons been mistaken for the diamond, owing to the lustre 
 of its crystals and its considerable hardness. It, however, 
 can always be distinguished from the diamond by its crys- 
 
 FIG. 150. 
 
 FIG. 148. 
 
 FIG. 149. 
 
 FIG. 151. 
 
 FIG. 152. 
 
 talline faces, hardness, and specific gravity (see example in 
 Table I.) 
 
 It usually occurs in six-sided prisms, more or less 
 modified, terminated with six-sided pyramids. Traces of 
 cleavage are seldom or ever apparent. Some of its salient 
 forms are shown in figs. 148, 149, 150, 151, 152, 153, 154. 
 
COLOURLESS STONES. 
 
 875 
 
 FIG. 153. 
 
 FIG. 154. 
 
 Some crystals are as pellucid as glass ; others, however, 
 assume all the shades of colour mentioned in the case of 
 the diamond. 
 
 Composition (Si0 2 ) : Pure silica or silicic acid. 
 
 WHITE ZIRCON. 
 
 Specific gravity, 4'44 to 4-8 ; hardness, 7'5. This stone 
 is often found crystallised in nature in four-sided prisms, 
 terminated by four-sided or rhomboidal or triangular 
 
 FIG. 156. 
 
 FIG. 157. 
 
 FIG. 155. 
 
 pyramids, and other forms. See figs. 155, 156, 157, 158, 
 159. 
 
 These stones are often employed in jewellery under the 
 name of ' rough diamonds.' They often occur brownish- 
 red and brown, red, yellow, and grey ; these varieties will 
 
876 GEMS AND PEECIOUS STONES. 
 
 be treated under their appropriate heads. It can be readily 
 distinguished from the diamond and quartz by hardness 
 
 FIG. 158. FIG. 160. 
 
 arid specific gravity ; also by the action of strong hydro- 
 chloric acid, which, if dropped on the diamond or quartz, 
 and allowed to remain for a little time, produces no change, 
 but if a zircon be so treated, the spot on which the acid 
 was placed remains dull. 
 
 Composition (Zr 2 3 ,Si0 2 ) : 
 
 Zirconia . 67'2 
 
 Silicic acid . . . . . . 33'5 
 
 100-7 
 WHITE SAPPHIRE. 
 
 Specific gravity, 3-97 to 4-27 ; hardness, 9. This stone, 
 in hardness, is next to the diamond. The sapphire occurs 
 variously coloured ; other colours will be discussed under 
 their appropriate heads. It crystallises in the rhombohe- 
 dric system, usually in six-sided prisms, but often so very 
 rough as not to be readily distinguishable. May be dis- 
 tinguished by gravity and hardness from all the preceding. 
 
 Composition (A1 2 3 ) : Pure alumina. 
 
 WHITE TOPAZ. 
 
 Specific gravity, 3'54 ; hardness, 9. This variety of 
 topaz, known for its limpidity by the term ' gouttes d'eau," 
 when polished has nearly the same lustre as the diamond ; 
 
COLOURLESS STONES. 
 
 877 
 
 the topaz, however, occurs of many colours see hereafter. 
 It crystallises in the right rectangular prismatic system. 
 Some of its natural forms are shown in figs. 161, 162, 163, 
 164, 165, 166. 
 
 It is readily rendered electric, and retains its electricity 
 for a very considerable time ; it is also pyro-electric, or 
 becomes electric when heated, a property by which it is 
 
 FIG. 161. FIG. 162. FIG. 163. 
 
 FIG. 164. 
 
 FIG. 165. 
 
 FIG. 166. 
 
 distinguished from the diamond, its specific gravity being 
 so similar that it cannot be made available as a means of 
 discriminating between the two stones. From the other 
 stones in this group, with the exception of the sapphire, 
 it is readily distinguished by its hardness and gravity, and 
 from the latter by its gravity and pyro-electricity. 
 Composition : 
 
 Silica 
 
 Alumina 
 
 Fluorine 
 
 34-2 
 57-5 
 
 7-8 
 
 99-5 
 
878 
 
 GEMS AND PRECIOUS STONES. 
 
 TABLE I. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF COLOURLESS STONES 
 WEIGHED IN AlR AND WATER. 
 
 Weight in 
 
 Weight in Water 
 
 Air 
 
 
 
 White 
 
 White 
 
 White 
 
 White 
 
 White 
 
 Grains 
 
 Zircon 
 
 Sapphire 
 
 Topaz 
 
 Diamond 
 
 Quartz 
 
 1 
 
 0-775 
 
 0-766 
 
 0-716 
 
 0-715 
 
 0-611 
 
 4 
 
 3-10 
 
 3-06 2-86 
 
 2-86 
 
 2-42 
 
 8 
 
 6-20 
 
 6-12 5-72 
 
 5-72 
 
 4-86 
 
 12 
 
 9-30 
 
 9-18 8-58 8-58 
 
 7-31 
 
 16 
 
 12-40 
 
 12-25 11-55 11-45 
 
 9-75 
 
 20 
 
 15-50 
 
 15-31 
 
 14-42 14-31 
 
 12-19 
 
 24 
 
 18-60 
 
 18-37 
 
 17-28 17-17 
 
 14-64 
 
 28 
 
 21-70 
 
 21-44 | 20-16 
 
 20-13 
 
 17-08 
 
 32 
 
 24-80 
 
 24-51 
 
 23-01 
 
 22-90 
 
 19-53 
 
 36 
 
 27-90 
 
 27-57 
 
 25-88 
 
 25-76 
 
 11-98 
 
 40 
 
 31-00 
 
 30-64 
 
 28-75 
 
 28-63 
 
 24-43 
 
 44 
 
 34-10 
 
 33-71 
 
 31-61 
 
 31-49 
 
 26-88 
 
 48 
 
 37-20 36-76 
 
 34-47 
 
 34-35 
 
 29-32 
 
 52 
 
 40-30 
 
 39-82 
 
 37-34 
 
 37-21 
 
 31-77 
 
 56 
 
 43-40 
 
 42-89 
 
 40-20 
 
 40-17 
 
 34-21 
 
 60 
 
 46-50 
 
 45-95 
 
 43-06 
 
 42-94 
 
 36-66 
 
 64 
 
 49-60 
 
 49-01 
 
 45-93 
 
 45-80 
 
 39-11 
 
 68 
 
 52-70 
 
 52-07 
 
 48-90 
 
 48-66 
 
 41-56 
 
 72 
 
 55-80 
 
 55-14 
 
 51-77 
 
 51-52 
 
 44-00 
 
 76 
 
 58-90 
 
 58-21 
 
 54-63 
 
 54-38 
 
 46-44 
 
 80 
 
 62-00 
 
 61-28 
 
 57-49 
 
 57-24 
 
 48-88 
 
 84 
 
 65-10 
 
 64-34 
 
 60-35 
 
 60-12 
 
 51-32 
 
 88 
 
 68-20 
 
 67-41 
 
 63-22 
 
 62-97 
 
 53-76 
 
 92 
 
 71-30 
 
 70-47 
 
 66-08 
 
 65-30 
 
 56-21 
 
 96 
 
 74-40 
 
 73-54 
 
 68-94 
 
 68-69 
 
 58-65 
 
 100 
 
 77-50 
 
 76-60 
 
 71-80 
 
 71-55 
 
 61-09 
 
 Specific 
 Gravity 
 
 \ 4-44 
 
 4-27 
 
 3-54 
 
 3-52 
 
 2-55 
 
 Example of the use of Table /.* A colourless stone, 
 weighing 40 grains in air, is reduced to 24-43 in water. 
 Look in the first column to 40, and then trace along its 
 horizontal line until a number very nearly approaching 
 24-43 is found ; refer then to the heading of the table, 
 above the number found, and the name there expressed 
 will be that of the stone examined. Supposing, however, 
 the weight of the stone be 41 grains, still the number 
 24-43 will be the nearest in the table, and -611 must be 
 added to it, as that sum would be the weight of 41 grains 
 
 * The Tables of Comparative Weights were calculated by Brard. 
 
YELLOW STONES. 879 
 
 of quartz or water. From the numbers obtained by cal- 
 culation also can the specific gravity be estimated. If 
 this course be pursued, refer to the bottom line of the 
 table for a corresponding number, and to the heading of the 
 table for the name of the stone. When the weight is any 
 even number of grains (that is, without fractions), the 
 readiest way is to refer to the table (first column) for the 
 number of grains, and then to the horizontal line to corre- 
 sponding number obtained, which is the weight in water. 
 
 Diamond and topaz, however, have very nearly equal 
 densities, and a second characteristic must be had recourse 
 to in order to determine the nature of two stones which 
 have an equal weight in water. This auxiliary character 
 is the development of electricity by heat, a phenomenon 
 exhibited by the topaz but not by the diamond. The test 
 of hardness may be also resorted to. 
 
 YELLOW STONES. 
 YELLOW ZIRCON ( JARGON). 
 
 The crystalline form, characteristics, and composition 
 of this stone have been described under the head ' White 
 Zircon.' 
 
 YELLOW SAPPHIRE. 
 
 Characteristics, &c., described under ' White Sapphire.' 
 
 CYMOPHANE (CHRYSO BERYL). 
 
 Specific gravity, 3'65 to 3-89 ; hardness, 8-5. The 
 cymophane is nearly as hard as the sapphire, harder than 
 the topaz and the emerald ; it readily scratches quartz. 
 Its colour is greenish-yellow, and has been placed in the 
 list of yellow stones rather than green, because usually the 
 yellowish tint is the most decided. This tint, which is very 
 agreeable in itself, is often relieved by a small spot of light 
 of a bluish- white tinge, which moves from point to point 
 of the stone as the position of the latter is varied. It is 
 
S80 
 
 GEMS AND PRECIOUS STONES. 
 
 rarely found in regular crystals, but more generally occurs 
 in rolled and rounded masses. For some of its iorms, 
 however, see figs. 167, 168, 169, 170. 
 
 FIG. 167. 
 
 FIG. 168. 
 
 FIG. 169. 
 
 FIG. 170. 
 
 Composition : No. 1 is a sample from the Brazils ; 
 No. 2, from Siberia. 
 
 i. 
 
 Alumina 78-10 
 
 Glucina 17-94 
 
 Iron oxide . . . . . 4*46 
 Chromium oxide .... 
 
 Copper and lead oxides ... 
 
 100-50 
 
 2. 
 
 78-92 
 
 18-02 
 
 3-12 
 
 0-36 
 
 0-29 
 
 100-71 
 
 YELLOW TOPAZ. 
 
 The general characteristics of this stone are described 
 under ' White Topaz.' 
 
 YELLOW TOURMALINE. 
 
 Specific gravity, 3-00 to 3-22 ; hardness, 7 to 7'5. The 
 tourmaline becomes electrical by heat ; one portion of a 
 crystal attracts light bodies, the other repels them. Its 
 
YELLOW STONES. 
 
 881 
 
 colour is very varied. The tourmaline has a vitreous 
 fracture. It occurs in semicrystalline prisms of irregular 
 
 FIG. 171. 
 
 FIG. 172. 
 
 FIG. 173. 
 
 FIG. 174. 
 
 FIG. 175. 
 
 FIG. 176. 
 
 form, generally deeply striated, and in prisms of six or 
 more sides, variously terminated, one end usually differing 
 from the other. 
 
 Figs. 171, 172, 173, 174, 175, and 176 represent some 
 of the forms of this mineral. 
 
 YELLOW EMEKALD. 
 
 Specific gravity, 273 to 2-76 ; hardness, 7-5 to 8. The 
 emerald occurs of many colours ; its tint par excellence is- 
 
 3 L 
 
882 
 
 GEMS AND PKECIOUS STONES. 
 
 green ; but there are many varieties tinged more or less 
 yellow or blue, and they even occur white. Its fracture 
 
 FIG. 178. 
 
 FIG. 177. 
 
 FIG. 179. 
 
 FIG. 180. 
 
 FIG. 181. 
 
 FIG. 183. 
 
 FIG. 182. 
 
 is vitreous, brilliant, and undulating. Its common form 
 is the hexahedral prism, sometimes deeply striated longi- 
 
YELLOW STONES. 
 
 883 
 
 tudinally. It readily cleaves parallel to all the planes of 
 its primary form the hexahedral prism. 
 
 The above are some of the forms it assumes : 
 178, 179, 180, 181, 182, and 183. 
 
 Composition : 
 
 figs. 17 
 
 Glucina 
 Silica 
 Alumina 
 Iron oxide 
 
 15-50 
 
 66-45 
 
 16-75 
 
 60 
 
 99-30 
 
 The 
 mium 
 
 ie green varieties contain a small quantity of chro- 
 oxide. 
 
 TABLE II. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF YELLOW STONES 
 WEIGHED IN AlE AND WATER. 
 
 Weight 
 in air 
 
 Weight in Water 
 
 Grrftins 
 
 Yellow 
 
 Yellow 
 
 Yellow 
 
 Yellow 
 
 Yellow 
 
 Yellow 
 
 Yellow 
 
 
 Zircon 
 
 Sapphire 
 
 Cymophane 
 
 Topaz 
 
 Tourmaline 
 
 Emerald 
 
 Quartz 
 
 i 
 
 0-775 
 
 0-766 
 
 9-738 
 
 0-716 
 
 0-690 
 
 0-633 
 
 0-611 
 
 4 
 
 3-10 
 
 3-06 
 
 2-95 
 
 2-86 
 
 2-76 
 
 2-53 
 
 2-42 
 
 8 
 
 6-20 
 
 6-12 
 
 5-90 
 
 5-72 
 
 5-52 
 
 5-06 
 
 4-86 
 
 12 
 
 9-30 
 
 9-18 
 
 8-85 
 
 8-58 
 
 8-28 
 
 7-59 
 
 7-31 
 
 16 
 
 12-40 
 
 12-25 
 
 11-80 
 
 11-55 
 
 11-04 
 
 10-12 
 
 9-75 
 
 20 
 
 15-50 
 
 15-31 
 
 14-75 
 
 14-42 
 
 13-80 
 
 12-65 
 
 12-19 
 
 24 
 
 18-60 
 
 18-07 
 
 17-70 
 
 17-28 
 
 16-56 
 
 15-19 
 
 14-04 
 
 28 
 
 21-70 
 
 21-44 
 
 20-65 
 
 20-15 
 
 19-32 
 
 17-72 
 
 17-08 
 
 32 
 
 24-80 
 
 24-51 
 
 23-60 
 
 23-01 
 
 20-08 
 
 20-25 
 
 19-53 
 
 36 
 
 27-90 
 
 27-57 
 
 26-55 
 
 25-88 
 
 24-84 
 
 22-77 
 
 21-98 
 
 40 
 
 31-00 
 
 30-64 
 
 29-50 
 
 29-75 
 
 27-60 
 
 25-30 
 
 24-43 
 
 44 
 
 34-10 
 
 33-71 
 
 32-45 
 
 31-61 
 
 30-36 
 
 27-83 
 
 26-88 
 
 48 
 
 37-20 
 
 36-76 
 
 35-40 
 
 34-47 
 
 33-12 
 
 30-36 
 
 29-32 
 
 52 
 
 40-30 
 
 39-82 
 
 38-35 
 
 37-34 
 
 35-88 
 
 32-89 
 
 31-77 
 
 56 
 
 43-40 
 
 42-89 
 
 41-30 
 
 40-20 
 
 38-64 
 
 35-43 
 
 34-21 
 
 60 
 
 46-50 
 
 45-95 
 
 44-25 
 
 43-06 
 
 41-40 
 
 37-94 
 
 36-66 
 
 64 
 
 49-60 
 
 49-01 
 
 47-20 
 
 45-93 
 
 44-16 
 
 40-47 
 
 39-11 
 
 68 
 
 52-70 
 
 52-08 
 
 50-15 
 
 48-90 
 
 46-92 
 
 43-00 
 
 41-56 
 
 72 
 
 55-80 
 
 55-14 
 
 53-10 
 
 51-77 
 
 49-68 
 
 45-53 
 
 44-00 
 
 76 
 
 58-90 
 
 58-21 
 
 56-05 
 
 54-63 
 
 52-44 
 
 48-07 
 
 46-44 
 
 80 
 
 62-00 
 
 61-28 
 
 59-00 
 
 57-49 
 
 55-20 
 
 50-60 
 
 48-88 
 
 84 
 
 65-10 
 
 64-34 
 
 61-95 
 
 60-35 
 
 57-96 
 
 53-13 
 
 51-32 
 
 88 
 
 68-20 
 
 67-41 
 
 64-90 
 
 63-22 
 
 60-72 
 
 55-66 
 
 53-76 
 
 92 
 
 71-30 
 
 70-47 
 
 67-85 
 
 66-08 
 
 63-48 
 
 58-19 
 
 56-21 
 
 96 
 
 74-40 
 
 73-54 
 
 70-80 
 
 68-94 
 
 66-24 
 
 60-72 
 
 58-65 
 
 100 
 
 77-50 
 
 76-60 
 
 73-75 
 
 71-80 
 
 69-00 
 
 63-25 
 
 61-09 
 
 Specific 
 Gravity 
 
 j- 4-44 
 
 4-27 
 
 3-89 
 
 3-53 
 
 3-22 
 
 2-72 
 
 2-55 
 
 3 L 2 
 
884 
 
 GEMS AND PRECIOUS STONES. 
 
 YELLOW QUARTZ. 
 
 For the characteristics, hardness, &c., of this mineral, 
 see ' White Quartz/ 
 
 BKOWN AND FLAME-COLOURED STONES. 
 
 ZIRCON (HYACINTH). 
 For characteristics, &c., see ' White Zircon/ 
 
 VERMEIL GARNET, NOBLE GARNET, ALMANDINE. 
 
 Specific gravity, 4 to 4*2 ; hardness, 6-5 to 7*5. There 
 are very many varieties of garnet, variously coloured ; but 
 
 FIG. 184. 
 
 FIG. 185. 
 
 FIG. 186 
 
 their crystalline form a rhombic dodecahedron, more 
 or less modified is a distinguishing characteristic. The 
 colouring matter of the garnet is iron. Figs. 184, 185, 186, 
 187, and 188 represent some of its crystalline forms. 
 
BROWN AND FLAME-COLOURED STONES. 
 
 FIG. 187. FIG. 188. 
 
 885 
 
 Composition : 
 
 Silica 
 
 Alumina . 
 Iron oxide 
 Manganese oxide 
 
 33-75 
 
 27-25 
 
 36-00 
 
 25 
 
 97-25 
 TABLE III. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF BROWNISH AND FLAME- 
 COLOURED STONES WEIGHED IN AIR AND WATER. 
 
 Weight in 
 Air 
 
 Weight in water 
 
 Grains 
 
 Hy acinthine Zircon 
 
 Vermeil Garnet 
 
 Essonite 
 
 Tourmaline 
 
 1 
 
 0-775 
 
 0-750 
 
 0-710 
 
 0-690 
 
 4 
 
 3-10 
 
 3-00 
 
 2-87 
 
 2-76 
 
 8 
 
 6-20 
 
 6-00 
 
 5-74 
 
 5-52 
 
 12 
 
 9-30 
 
 9-00 
 
 8-61 
 
 8-28 
 
 16 
 
 12-40 
 
 12-00 
 
 11-48 
 
 11-04 
 
 20 
 
 15-50 
 
 15-00 
 
 14-35 
 
 13-80 
 
 24 
 
 18-60 
 
 18-00 
 
 17-22 
 
 16-56 
 
 28 
 
 21-70 
 
 21-00 
 
 20-09 
 
 19-32 
 
 32 
 
 24-80 
 
 24-00 
 
 22-96 
 
 22-08 
 
 36 
 
 27-90 
 
 27-00 
 
 25-83 
 
 24-84 
 
 40 
 
 31-30 
 
 30-00 
 
 28-70 
 
 27-60 
 
 44 
 
 34-10 
 
 33-00 
 
 31-57 
 
 30-36 
 
 48 
 
 37-20 
 
 36-00 
 
 34-44 
 
 33-12 
 
 52 
 
 40-30 
 
 39-00 
 
 37-31 
 
 35-88 
 
 56 
 
 43-40 
 
 42-00 
 
 40-18 
 
 38-64 
 
 60 
 
 46-50 
 
 45-00 
 
 43-05 
 
 41-40 
 
 64 
 
 49-60 
 
 48-00 
 
 45-92 
 
 44-16 
 
 68 
 
 52-70 
 
 51-00 
 
 48-79 
 
 46-92 
 
 72 
 
 55-80 
 
 54-00 
 
 51-66 
 
 49-68 
 
 76 
 
 58-90 
 
 57-00 
 
 54-53 
 
 52-44 
 
 80 
 
 61-00 
 
 60-00 
 
 57-40 
 
 55-20 
 
 84 
 
 65-10 
 
 63-00 
 
 60-27 
 
 57-96 
 
 88 
 
 68-20 
 
 66-00 
 
 63-14 
 
 60-72 
 
 92 
 
 71-30 
 
 69-00 
 
 66-01 
 
 63-48 
 
 96 
 
 74-40 
 
 72-00 
 
 68-88 
 
 66-24 
 
 100 
 
 77-50 
 
 75-00 
 
 71-75 
 
 69-00 
 
 Specific 
 
 Gravity 
 
 } 4-44 
 
 4-00 
 
 3-54 
 
 3-22 
 
886 GEMS AND PKECIOUS STONES. 
 
 ESSONITE, CINNAMON STONE. 
 
 Specific gravity, 3*5 to 3 -6. This stone has an agree- 
 able orange-yellow tinge, which becomes a warm and 
 brilliant tint when 'the mass is large. This stone is not 
 usually found crystalline, but in irregular forms and 
 masses, which are characterised by fissures in all direc- 
 tions. 
 
 Composition : 
 
 Silica 38-80 
 
 Alumina 21-20 
 
 Lime 31-25 
 
 Iron oxide with small quantities of Potash "1 Q-Q 
 and Magnesia / 
 
 97-75 
 TOURMALINE. 
 
 For the characteristics of this mineral see 'Yellow 
 Tourmaline.' 
 
 KED AND KOSE-COLOURED STONES. 
 EED SAPPHIEE (ORIENTAL RUBY). 
 
 For characteristics, crystalline form, &c., see ' White 
 Sapphire.' 
 
 DEEP RED GARNET, NOBLE GARNET. 
 
 For characteristics, &c., see ' Vermeil Garnet.' 
 
 SPINEL RUBY. 
 
 Specific gravity, 3*5 to 3-6 ; hardness, 8. The spinel 
 readily scratches quartz, but is scratched by the sapphire. 
 Its special colour is red, approaching a rose tint : this 
 tinge, however, undergoes various modifications, such as 
 scarlet, red, rose, yellowish-red, and reddish-purple : it 
 is also found blue and black. Its fracture is flattish- 
 conchoidal, with a splendent vitreous lustre. It occurs 
 crystallised in regular octahedrons, sometimes having 
 their edges replaced as in macles : sometimes it assumes 
 the globular form. The spinel may be distinguished from 
 the true ruby and the garnet by hardness and specific 
 
EED AND KOSE-COLOUKED STONES. 
 
 887 
 
 gravity ; and from reddish topaz, which possesses nearly 
 the same specific gravity, by its electric properties. 
 Composition of spinel ruby : 
 
 Silica -02 
 
 Alumina . 69-01 
 
 Magnesia 26'21 
 
 Iron protoxide 0'71 
 
 Chromium oxide . . . . . 1-11 
 
 99-06 
 EEDDISH TOPAZ. 
 
 For characteristics, &c., see ' White Topaz.' 
 
 KED TOURMALINE. 
 
 For characteristics, &c., see 4 Yellow Tourmaline.' 
 
 TABLE IV. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF EED OR EOSE-COLOURED 
 STONES WEIGHED IN AIR AND WATER. 
 
 Weight 
 
 Weight in Water 
 
 in Air 
 
 
 Grains 
 
 Red Sapphire 
 (True Ruby) 
 
 Deep Garnets 
 
 Spinel 
 
 Smoke or Red 
 Topaz 
 
 Red Tourmaline 
 
 1 
 
 0-766 
 
 0-750 
 
 0-722 
 
 0-716 
 
 0-690 
 
 4 
 
 3-060 
 
 3-700 
 
 2-880 
 
 2-860 
 
 2-760 
 
 8 
 
 6-120 
 
 6-000 
 
 5-770 
 
 5-720 
 
 5-520 
 
 12 
 
 9-180 
 
 9-000 
 
 8-660 
 
 8-585 
 
 8-280 
 
 16 
 
 12-250 
 
 12-000 
 
 11-550 
 
 11-550 
 
 11-040 
 
 20 
 
 15-310 
 
 15-000 
 
 14-440 
 
 14-420 
 
 13-800 
 
 24 
 
 18-370 
 
 18-000 
 
 17-330 
 
 17-280 
 
 16-560 
 
 28 
 
 21-440 
 
 21-000 
 
 20-220 
 
 20-150 
 
 19-320 
 
 32 
 
 24-510 
 
 24-000 
 
 23-110 
 
 23-610 
 
 22-080 
 
 36 
 
 27-570 
 
 27-000 
 
 26-000 
 
 25-880 
 
 24-840 
 
 40 
 
 30-640 
 
 30-000 
 
 28-880 
 
 28-750 
 
 27-600 
 
 44 
 
 33-710 
 
 33-000 
 
 31-770 
 
 31-610 
 
 30-360 
 
 48 
 
 36-760 
 
 36-000 
 
 34-660 
 
 34-470 
 
 33-120 
 
 52 
 
 39-820 
 
 39-000 
 
 37-550 
 
 37-340 
 
 35-880 
 
 56 
 
 42-890 
 
 42-000 
 
 40-440 
 
 40-200 
 
 38-640 
 
 60 
 
 44-950 
 
 45-000 
 
 43-300 
 
 43-060 
 
 41-400 
 
 64 
 
 49-010 
 
 48-000 
 
 46-220 
 
 45-930 
 
 44-160 
 
 68 
 
 52-080 
 
 51-000 
 
 49-110 
 
 48-900 
 
 46-920 
 
 72 
 
 55-140 
 
 54-000 
 
 51-990 
 
 51-770 
 
 49-680 
 
 76 
 
 58-210 
 
 57-000 
 
 54-880 
 
 54-630 
 
 52-440 
 
 80 
 
 61-280 
 
 60-000 
 
 57-770 
 
 57-490 
 
 52-200 
 
 84 
 
 64-340 
 
 63-000 
 
 60-660 
 
 60-350 
 
 57-960 
 
 88 
 
 67-410 
 
 66-000 
 
 63-550 
 
 63-220 
 
 60-720 
 
 92 
 
 70-470 
 
 69-000 
 
 66-440 
 
 66-080 
 
 63-480 
 
 96 
 
 73-540 
 
 72-000 
 
 69-330 
 
 68-940 
 
 66-240 
 
 100 
 
 76-600 
 
 75-000 
 
 72-220 
 
 71-800 
 
 69-000 
 
 Specific 
 Gravity 
 
 } 4-270 
 
 4-000 
 
 3-600 
 
 3-530 
 
 3-220 
 
888 GEMS AND PRECIOUS STONES. 
 
 BLUE STONES. 
 BLUE SAPPHIEE. 
 
 For characteristics, &c., see ' White Sapphire.' 
 
 DISTHENE, CYANITE. 
 
 Specific gravity, 3'5 to 3*7 ; hardness, 5 to 7. Fine 
 specimens of disthene possess a bright blue colour, which 
 passes insensibly into a deep sky blue. Its transparency 
 is nearly perfect, and it presents small pearly reflections, 
 which add to the beauty of its colour. The primary form 
 of its crystals is a doubly oblique prism, and they cleave 
 very readily in the direction of their length. It can be 
 
 FIG. 189. 
 
 FIG. 190, FIG. 191. 
 
 readily distinguished from the sapphire by its being less 
 hard, as also by its specific gravity. Figs. 189, 190, and 
 191 represent some of its crystalline forms. 
 
 Composition of a specimen from St. Gothard : 
 
 Silica 43-0 
 
 Alumina 55-0 
 
 Iron oxide -5 
 
 98-5 
 BLUE TOPAZ. 
 
 For characteristics, &c., see fc White Topaz.' Blue 
 topaz and disthene, having the same specific gravity, may 
 by that test alone be confounded with each other ; but 
 the appearance of each is so different that they can 
 rarely be confounded. If, however, the electrical test be 
 applied, no fear of mistaking one for the other need be 
 entertained, as only the topaz becomes electrical. 
 
BLUE STONES. 
 BLUE TOURMALINE. 
 
 For characteristics, &c., see ' Yellow Tourmaline.' 
 
 BLUE BERYL. 
 
 For characteristics, &c., see 'Emerald.' The tint and 
 appearance of this stone and that of the blue topaz are 
 so similar that they cannot be distinguished by that test ; 
 their specific gravities, however, are so different that they 
 may, by this simple means, be readily discriminated. 
 
 DICHROITE, WATER SAPPHIRE. 
 
 Specific gravity, 2*56 to 2-65 ; hardness, 7 to 7'5. The 
 chief characteristic of this stone is that it possesses a 
 double colour ; that is, it is a fine blue or a normal yellow, 
 as it is viewed in the direction of its base, or the planes 
 of a hexahedral prism, which is its crystalline form. It 
 can be thus readily distinguished, as also by its having 
 nearly the same specific gravity as quartz, and thus being 
 the lightest of the blue stones. Composition : 
 
 Silica. . . 
 Alumina 
 Magnesia . 
 Iron protoxide . 
 Manganese protoxide 
 Loss in fire (water ? ) 
 
 48-35 
 
 31-71 
 
 10-16 
 
 8-12 
 
 33 
 
 60 
 
 99-27 
 
 TURQUOISE. 
 
 Specific gravity, 2-8 to 3 ; hardness, 5 to 6. This stone 
 has not been placed in the list of specific gravities, as it 
 can be so readily detected by its appearance. It is bright 
 or greenish-blue in colour ; its aspect is earthy or compact. 
 It scratches apatite, and even glass ; but is scratched by 
 quartz. It occurs filling fissures, or forming concretions 
 in siliceous and argillo-ferruginous rocks. Composition : 
 
 Phosphoric acid 17-86 
 
 Alumina . 
 
 Silica 
 
 Iron peroxide . 
 
 Lime . 
 
 Water and fluoric acid 
 
 10-01 
 8-90 
 
 36-82 
 0-15 
 
 25-95 
 
 99-69 
 
890 
 
 GEMS AND PRECIOUS STONES. 
 
 TABLE V. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF BLUE STONES WEIGHED 
 IN AIR AND WATER. 
 
 
 Weight iii Water 
 
 Weight 
 
 
 in Air 
 
 
 
 
 
 | 
 
 Grains 
 
 Blue 
 Sapphire 
 
 Disthene 
 Cyanite 
 
 Blue 
 Topaz 
 
 Tourmaline 
 
 Blue 
 Beryl 
 
 Dichroite, 
 Water 
 Sapphire 
 
 1 
 
 0-766 
 
 0-717 
 
 0-716 
 
 0-690 
 
 0-633 
 
 0-622 
 
 4 
 
 3-06 
 
 2-87 
 
 2-86 
 
 2-16 
 
 2-53 
 
 2-49 
 
 8 
 
 6-12 
 
 5-74 
 
 5-72 
 
 5-52 
 
 5-06 
 
 4-98 
 
 12 
 
 9-18 
 
 8-61 
 
 8-58 
 
 8-28 
 
 7-59 
 
 7-47 
 
 16 
 
 12-25 
 
 11-48 
 
 11-45 11-04 
 
 10-12 
 
 9-96 
 
 20 
 
 15-31 
 
 ' 14-35 
 
 14-42 
 
 13-80 
 
 12-65 
 
 12-45 
 
 24 
 
 18-37 
 
 17-22 
 
 17-18 
 
 16-56 
 
 15-19 
 
 14-94 
 
 28 
 
 21-44 
 
 20-09 
 
 20-05 
 
 19-32 
 
 17-72 
 
 17-43 
 
 82 
 
 24-51 
 
 22-96 
 
 22-91 
 
 20-08 
 
 20-25 
 
 19-92 
 
 36 
 
 27-57 
 
 25-83 
 
 25-78 
 
 24-84 
 
 22-77 
 
 22-41 
 
 40 
 
 30-64 
 
 28-70 
 
 28-65 
 
 27-60 
 
 25-30 
 
 24-90 
 
 44 
 
 33-71 
 
 31-57 31-51 
 
 30-66 
 
 27-83 
 
 27-39 
 
 48 
 
 36-76 
 
 34-44 34-37 
 
 33-12 
 
 30-36 
 
 29-88 
 
 52 
 
 39-82 
 
 37-31 37-24 
 
 35-88 
 
 32-89 
 
 32-37 
 
 56 
 
 42-89 
 
 40-18 40-10 
 
 38-64 
 
 35-43 
 
 34-88 
 
 60 
 
 45-95 
 
 43-05 42-96 
 
 41-40 
 
 37-94 
 
 37-35 
 
 64 
 
 49-01 
 
 45-92 45-83 
 
 44-16 
 
 40-47 
 
 39-84 
 
 68 
 
 52-08 
 
 48-79 48-80 
 
 46-92 
 
 43-00 
 
 42-33 
 
 72 
 
 55-14 
 
 51-66 51-67 
 
 49-68 
 
 45-53 
 
 44-82 
 
 76 
 
 58-21 
 
 54-53 54-53 
 
 52-44 
 
 48-07 
 
 47-31 
 
 80 
 
 61-28 
 
 57-40 57-49 
 
 55-20 
 
 50-60 
 
 49-80 
 
 84 
 
 64-34 
 
 60-27 60-25 
 
 57-96 
 
 53-13 
 
 52-29 
 
 88 
 
 67-41 
 
 63-14 63-12 
 
 60-72 
 
 55-66 
 
 54-78 
 
 92 
 
 70-47 
 
 66-01 
 
 65-98 
 
 63-48 
 
 58-19 
 
 57-27 
 
 96 
 
 73-54 
 
 68-88 
 
 63-84 
 
 66-24 
 
 60-72 
 
 59-76 
 
 100 
 
 76-60 
 
 71-75 
 
 71-70 
 
 69-00 
 
 63-25 
 
 62-25 
 
 Specific 
 Gravity 
 
 | 4-27 
 
 3-54 
 
 3-53 
 
 3-22 
 
 2-72 
 
 2-65 
 
 VIOLET STONES. 
 VIOLET SAPPHIRE. 
 
 For characteristics, &c., see 4 White Sapphire.' 
 
 VIOLET TOURMALINE. 
 
 For characteristics, &c., see ' Yellow Tourmaline.' 
 
 VIOLET QUARTZ, AMETHYST. 
 
 For characteristics, &c., see ' White Quartz.' 
 
GREEN STONES. 
 
 891 
 
 TABLE VI. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF VIOLET STONES WEIGHED 
 IN AIR AND WATER. 
 
 
 Weight in Air 
 
 Weight in Air 
 
 
 Grains 
 
 Violet Sapphire 
 
 Violet Tourmaline 
 
 Amethystine Quartz 
 (Amethyst; 
 
 1 
 
 0-766 
 
 0-690 
 
 0-611 
 
 4 
 
 3-06 
 
 2-76 
 
 2-42 
 
 8 
 
 6-12 
 
 5-52 
 
 4-86 
 
 12 
 
 9-18 
 
 8-28 
 
 7-31 
 
 16 
 
 12-25 
 
 11-04 
 
 9-75 
 
 20 
 
 15-31 
 
 13-80 
 
 12-19 
 
 24 
 
 18-37 
 
 16-56 
 
 14-64 
 
 28 
 
 21-44 
 
 19-32 
 
 17-08 
 
 32 
 
 24-51 
 
 20-08 
 
 19-53 
 
 36 
 
 27-57 
 
 24-84 
 
 21-98 
 
 40 
 
 30-64 
 
 27-60 
 
 24-43 
 
 44 
 
 33-71 
 
 30-36 
 
 26-88 
 
 48 
 
 36-76 
 
 33-12 
 
 29-32 
 
 52 
 
 39-82 
 
 35-88 
 
 31-77 
 
 56 
 
 42-89 
 
 38-64 
 
 34-21 
 
 60 
 
 45-95 
 
 41-40 
 
 36-66 
 
 64 
 
 49-01 
 
 44-16 
 
 39-11 
 
 68 
 
 52-02 
 
 46-92 
 
 41-56 
 
 72 
 
 55-14 
 
 49-68 
 
 44-00 
 
 76 
 
 58-21 
 
 52-44 
 
 46-44 
 
 80 
 
 61-28 
 
 55-20 
 
 48-88 
 
 84 
 
 64-34 
 
 57-96 
 
 51-32 
 
 88 
 
 67-41 
 
 60-72 
 
 53-76 
 
 92 
 
 70-47 
 
 63-48 
 
 56-21 
 
 96 
 
 73-54 
 
 66-24 
 
 58-65 
 
 100 
 
 76-60 
 
 69-00 
 
 61-09 
 
 Specific Gravity 
 
 4-27 
 
 3-22 
 
 2-55 
 
 GBEEN STONES. 
 GREEN SAPPHIRE. 
 
 For characteristics, &c., see 'Yellow Emerald.' 
 
 PERIDOT, CRYSOLITE. 
 
 Specific gravity, 3*3 to 3-5 ; hardness, 6*5 to 7.. This 
 stone has a more or less deep olive or yellowish -green 
 colour. It is more generally found in rolled grains than 
 in regular prismatic crystals. It is possessed in a very 
 high degree of double refraction. Figs. 192, 193, 194, 
 and 195 represent some of its crystalline forms. 
 
392 
 
 GEMS AND PEECIOUS STONES. 
 
 GEEEN TOUEMALINE. 
 
 For characteristics, see ' Yellow Tourmaline. 
 
 EMEEALD. 
 
 For characteristics, see ' Yellow Emerald.' 
 
 FIG. 192. 
 
 FIG. 193. 
 
 FIG. 194. 
 
 
 FIG. 195. 
 
 AQUA-MAEINE. 
 
 This stone possesses a very pale green tinge. For 
 other characteristics, see ' Yellow Emerald.' 
 
 CHEYSOPEASE. 
 
 This mineral is a green-coloured quartz, and can be 
 readily recognised by referring to the characteristics of 
 quartz. 
 
GREEN STONES. 
 
 TABLE VII. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF GREEN STONES WEIGHED 
 IN AIR AND WATER. 
 
 Weight 
 
 1T1 A ! 
 
 Weight in Water 
 
 in Air 
 Grains 
 
 Green 
 Sapphire 
 
 Peridot 
 
 Green 
 Tourmaline 
 
 Emerald 
 
 Aqua-marine 
 
 Chrysoprase 
 
 i 
 
 0-766 
 
 0-708 
 
 0-690 
 
 0-633 
 
 0-633 
 
 0-611 
 
 4 
 
 3-06 
 
 2-83 
 
 2-76 
 
 2-53 
 
 2-53 
 
 2-42 
 
 8 
 
 6-12 
 
 5-66 
 
 5-52 
 
 5-06 
 
 5-06 
 
 4-86 
 
 12 
 
 9-18 
 
 8-49 
 
 8-28 
 
 7-59 
 
 7-59 
 
 7-31 
 
 16 
 
 12-25 
 
 11-32 
 
 11-04 
 
 10-12 
 
 10-12 
 
 9-75 
 
 20 
 
 15-31 
 
 14-16 
 
 13-18 
 
 12-65 
 
 12-65 
 
 12-19 
 
 24 
 
 18-37 
 
 16-99 
 
 16-56 
 
 15-19 
 
 15-19 
 
 14-64 
 
 28 
 
 21-44 
 
 19-82 
 
 19-32 
 
 17-72 
 
 17-72 
 
 17-08 
 
 32 
 
 24-51 
 
 22-65 
 
 22-08 20-25 
 
 20-25 
 
 19-53 
 
 36 
 
 27-57 
 
 25-48 
 
 24-84 
 
 22-77 
 
 27-77 
 
 21-98 
 
 40 
 
 30-64 
 
 28-32 
 
 27-60 
 
 25-30 
 
 25-30 
 
 24-43 
 
 44 
 
 33-71 
 
 31-15 30-36 
 
 27-83 
 
 27-83 
 
 36-88 
 
 48 
 
 36-76 
 
 33-98 
 
 33-12 
 
 30-36 
 
 30-36 
 
 29-32 
 
 52 
 
 39-82 
 
 36-81 
 
 35-88 32-89 
 
 32-89 
 
 31-77 
 
 56 
 
 42-89 
 
 39-64 
 
 38-64 
 
 35-43 
 
 35-43 
 
 34-21 
 
 60 
 
 45-95 
 
 42-48 
 
 41-40 
 
 37-94 
 
 37-94 
 
 36-66 
 
 64 
 
 49-01 
 
 45-31 
 
 44-16 
 
 40-47 
 
 40-47 
 
 39-11 
 
 68 
 
 52-08 
 
 48-14 
 
 46-92 
 
 43-00 
 
 43-00 
 
 41-56 
 
 72 
 
 55-14 
 
 50-97 
 
 49-68 
 
 45-53 
 
 45-53 
 
 44-00 
 
 76 
 
 58-21 
 
 53-80 
 
 52-44 
 
 48-07 
 
 48-07 
 
 46-44 
 
 80 
 
 61-28 
 
 56-64 
 
 55-20 
 
 50-60 
 
 50-60 
 
 48-88 
 
 84 
 
 64-34 
 
 59-47 
 
 57-96 
 
 53-13 
 
 53-13 
 
 51-32 
 
 88 
 
 67-41 
 
 62-30 
 
 60-72 
 
 55-66 
 
 55-66 
 
 53-76 
 
 92 
 
 70-47 
 
 65-13 
 
 63-48 
 
 58-19 
 
 58-19 
 
 56-21 
 
 96 
 
 73-54 
 
 67-96 
 
 66-24 
 
 60-72 
 
 60-72 
 
 58-65 
 
 100 
 
 76-60 
 
 70-80 
 
 69-00 
 
 63-25 
 
 63-25 
 
 61-09 
 
 Specific 
 Gravity 
 
 | 4-27 
 
 3-42 
 
 3-22 
 
 2-72 
 
 2-72 
 
 2-56 
 
 STONES POSSESSING A PLAY OF COLOUES 
 (CHATOYANT). 
 
 In the following list of stones no regard has been paid 
 to absolute colours, but only to the play of colours the 
 stones exhibit. This play or reflection is of two kinds : 
 in some, as the sapphires, it appears as a white star with 
 six rays, on a blue, red, or yellow ground ; or on a purple 
 ground in the garnet. In others it is but a point or mass 
 of pearly light, which sometimes appears to occupy the 
 whole of the stone, and varies according to the inclination 
 given to the stone. The cymophane, cry soli te quartz, 
 Egyptian emerald, felspar, and cat's eye belong to this 
 class. 
 
GEMS AND PRECIOUS STONES. 
 
 The specific gravities of such stones as the opal, &c., 
 have not been given, as their appearance sufficiently charac- 
 terises them. 
 
 SAPPHIRE. 
 
 For characteristics, &c., see 'White Sapphire.' 
 
 GARNET. 
 
 For characteristics, &c., see ' Vermeil Garnet.' 
 
 See ' Cymophane.' 
 
 CYMOPHANE. 
 
 ANTIQUE EMERALD. 
 
 For characteristics, &c., see ' Yellow Emerald.' 
 
 QUARTZ. 
 
 See ' White Quartz.' 
 
 FIG. 196. FIG. 197. FIG. 198. 
 
 FIG. 199. 
 
 FIG. 201. 
 
 FIG. 200. 
 
STONES POSSESSING A PLAY OF COLOURS. 
 
 895 
 
 FELSPAR, NACREOUS FELSPAR, FISH-EYE, ETC. 
 
 Specific gravity, 2'3 to 2-5 ; hardness 4'5 to 5. This 
 species of felspar has a lamellar texture. It will be seen 
 by the lowness of its specific gravity that it cannot be 
 readily confounded with other stones. In appearance its 
 transparency is nebulous, and it presents pearly white 
 reflections, which float about and vacillate in proportion 
 as its position changes. The foregoing are some of the 
 forms of felspar ; see figs. 183, 184, 185, 186, 187, and 188. 
 
 Composition : 
 
 Potash 5-26 
 
 Silica 52-90 
 
 Lime 25-20 
 
 Water ....... 16-00 
 
 Fluoric acid 0*82 
 
 100-18 
 TABLE VIII. 
 
 COMPARATIVE TABLE OF THE WEIGHTS OF STONES POSSESSING A 
 PLAY OF COLOURS (CHATOYANT). 
 
 
 Weight in Water 
 
 Weight 
 
 
 in Air 
 Grains 
 
 Sapphires 
 
 Garnets 
 
 Gymophane 
 
 Antique 
 Emerald 
 
 Quartz 
 
 Felspar 
 
 1 
 
 0-766 
 
 0-750 
 
 0-738 
 
 0-633 
 
 0-611 
 
 0-592 
 
 4 
 
 3-06 
 
 3-00 
 
 2-95 
 
 2-53 
 
 2-42 
 
 2-37 
 
 8 
 
 6-12 
 
 6-00 
 
 5-90 
 
 5-06 
 
 4-86 
 
 4-74 
 
 12 
 
 9-18 
 
 9-00 
 
 8-85 
 
 7-59 
 
 7-31 
 
 7-11 
 
 16 
 
 12-25 
 
 12-00 
 
 11-80 
 
 10-12 
 
 9-75 
 
 9-47 
 
 20 
 
 15-31 
 
 15-00 
 
 14-75 
 
 12-65 
 
 12-19 
 
 11-84 
 
 24 
 
 18-37 
 
 18-00 
 
 17-70 
 
 15-19 
 
 14-64 
 
 14-20 
 
 28 
 
 21-44 
 
 21-00 
 
 20-65 
 
 17-72 
 
 17-08 
 
 16-57 
 
 32 
 
 24-51 
 
 24-00 
 
 23-60 
 
 20-25 
 
 19-53 
 
 18-94 
 
 36 
 
 27-57 
 
 27-00 
 
 26-55 
 
 22-77 
 
 21-98 
 
 21-31 
 
 40 
 
 30-64 
 
 30-00 
 
 29-50 
 
 25-30 
 
 24-43 
 
 23-68 
 
 44 
 
 33-71 
 
 33-00 
 
 32-46 
 
 27-83 
 
 26-88 
 
 26-05 
 
 48 
 
 36-76 
 
 36-00 
 
 35-40 
 
 30-36 
 
 29-32 
 
 28-42 
 
 52 
 
 39-84 
 
 39-00 
 
 38-35 
 
 32-89 
 
 31-77 
 
 30-79 
 
 56 
 
 42-89 
 
 42-00 
 
 41-30 
 
 35-43 
 
 34-21 
 
 33-15 
 
 60 
 
 45-95 
 
 45-00 
 
 44-25 
 
 37-94 
 
 36-66 
 
 35-52 
 
 64 
 
 49-01 
 
 48-00 
 
 47-20 
 
 40-47 
 
 39-11 
 
 37-88 
 
 68 
 
 52-07 
 
 51-00 
 
 50-15 
 
 43-00 
 
 41-56 
 
 40-25 
 
 72 
 
 55-14 
 
 54-00 
 
 53-10 
 
 45-53 
 
 44-00 
 
 42-62 
 
 76 
 
 58-21 
 
 57-00 
 
 56-05 
 
 48-07 
 
 46-44 
 
 44-99 
 
 80 
 
 61-28 
 
 60-00 
 
 59-00 
 
 50-60 
 
 48-88 
 
 47-36 
 
 84 
 
 64-34 
 
 63-00 
 
 61-95 
 
 53-13 
 
 51-32 
 
 49-73 
 
 88 
 
 67-47 
 
 66-00 
 
 64-90 
 
 55-66 
 
 53-76 
 
 52-10 
 
 92 
 
 70-47 
 
 69-00 
 
 67-85 
 
 58-19 
 
 56-21 
 
 54-47 
 
 96 
 
 73-54 
 
 72-00 
 
 70-80 
 
 60-72 
 
 58-65 
 
 56-84 
 
 100 
 
 76-60 
 
 75-00 
 
 73-75 
 
 63-25 
 
 61-09 
 
 59-21 
 
 Specific 
 Gravity 
 
 | 4-27 
 
 4-00 
 
 3-89 
 
 2-72 
 
 2-55 
 
 2-45 
 
896 GEMS AND PRECIOUS STONES. 
 
 GLASS AND AETIFICIAL GEMS. 
 
 Glass is often used to imitate gems, but can be easily 
 distinguished by the following characters : 
 
 1. Inferior brilliancy. Though in many cases artificial 
 gems have a fine lustre, they are invariably soft. The 
 materials which communicate brilliancy to glass impair 
 its hardness ; and the result is that glass gems, when 
 examined by a lens, are generally found to have blunt or 
 jagged corners and edges, and surfaces covered with 
 minute, irregular scratches. This is invariably the case 
 after the glass gem has been a little in use, and thus the 
 brilliancy soon becomes impaired. 
 
 2. Inferior hardness. Artificial gems can be scratched 
 with a knife, using a slight pressure. Faint scratches are 
 made visible by breathing upon them, whereas true gems 
 retain the original brilliancy which they possessed on 
 leaving the lapidary's wheel. The polished faces of true 
 gems cannot be scratched with a knife ; moreover, after 
 long wear, they show no signs of scratches. 
 
 3. Polished gems become readily electric by friction, 
 particularly the topaz and diamond ; but glass imitations 
 require much longer friction to produce the same effect, 
 and also retain the electric power for a shorter time. 
 
 4. Glass is fusible in the blowpipe flame. 
 
APPENDIX. 
 
TABLE L 
 
 Showing the Quantity of FINE GOLD in 1 oz. of any ALLOY to 
 of a Carat Grain and the MINT VALUE of 1 oz. of each 
 Alloy. 
 
 FINE GOLD, 
 
 Per Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Drvts. Grs. 
 
 Carats 
 
 GTS. E 
 
 'sjhtlis 
 
 
 
 s. 
 
 d. 
 
 1 
 
 
 
 
 
 000 
 
 24 
 
 
 
 
 
 4 
 
 4 
 
 11-4545 
 
 
 
 19 
 
 23 
 
 375 
 
 23 
 
 3 
 
 7 
 
 4 
 
 4 
 
 10-1271 
 
 
 
 19 
 
 22 
 
 750 
 
 23 
 
 3 
 
 6 
 
 4 
 
 4 
 
 8-7997 
 
 
 
 19 
 
 22 
 
 125 
 
 23 
 
 3 
 
 5 
 
 4 
 
 4 
 
 7-4723 
 
 
 
 19 
 
 21 
 
 500 
 
 23 
 
 3 
 
 4 
 
 4 
 
 4 
 
 6-1448 
 
 
 
 19 
 
 20 
 
 875 
 
 23 
 
 3 
 
 3 
 
 4 
 
 4 
 
 4-8174 
 
 
 
 19 
 
 20 
 
 250 
 
 23 
 
 3 
 
 2 
 
 4 
 
 4 
 
 3-4900 
 
 
 
 19 
 
 19 
 
 625 
 
 23 
 
 3 
 
 1 
 
 4 
 
 4 
 
 2-1626 
 
 
 
 19 
 
 19 
 
 000 
 
 23 
 
 3 
 
 
 
 4 
 
 4 
 
 0-8352 
 
 
 
 19 
 
 18 
 
 375 
 
 23 
 
 2 
 
 7 
 
 4 
 
 3 
 
 11-5078 
 
 
 
 19 
 
 17 
 
 750 
 
 23 
 
 2 
 
 6 
 
 4 
 
 3 
 
 10-1804 
 
 
 
 19 
 
 17 
 
 125 
 
 23 
 
 2 
 
 5 
 
 4 
 
 3 
 
 8-8529 
 
 
 
 19 
 
 16 
 
 500 
 
 23 
 
 2 
 
 4 
 
 4 
 
 3 
 
 7-5255 
 
 
 
 19 
 
 15 
 
 875 
 
 23 
 
 2 
 
 3 
 
 4 
 
 3 
 
 6-1981 
 
 
 
 19 
 
 15 
 
 250 
 
 23 
 
 2 
 
 2 
 
 4 
 
 3 
 
 4-8707 
 
 
 
 19 
 
 14 
 
 625 
 
 23 
 
 2 
 
 1 
 
 4 
 
 3 
 
 3-5433 
 
 
 
 19 
 
 14 
 
 000 
 
 23 
 
 2 
 
 
 
 4 
 
 3 
 
 2-2159 
 
 .0 
 
 19 
 
 13 
 
 375 
 
 23 
 
 1 
 
 7 
 
 4 
 
 3 
 
 0-8885 
 
 
 
 19 
 
 12 
 
 750 
 
 23 
 
 1 
 
 6 
 
 4 
 
 2 
 
 11-5610 
 
 
 
 19 
 
 12 
 
 125 
 
 23 
 
 1 
 
 5 
 
 4 
 
 2 
 
 10-2336 
 
 
 
 19 
 
 11 
 
 500 
 
 23 
 
 1 
 
 4 
 
 4 
 
 2 
 
 8-9062 
 
 
 
 19 
 
 10 
 
 875 
 
 23 
 
 1 
 
 3 
 
 4 
 
 2 
 
 7-5788 
 
 
 
 19 
 
 10 
 
 250 
 
 23 
 
 1 
 
 2 
 
 4 
 
 2 
 
 6-2514 
 
 
 
 19 
 
 9 
 
 625 
 
 23 
 
 1 
 
 1 
 
 4 
 
 2 
 
 4-9240 
 
 
 
 19 
 
 9 
 
 000 
 
 23 
 
 1 
 
 
 
 4 
 
 2 
 
 3-5965 
 
 
 
 19 
 
 8 
 
 375 
 
 23 
 
 
 
 7 
 
 4 
 
 2 
 
 2-2691 
 
 
 
 19 
 
 7 
 
 750 
 
 23 
 
 
 
 6 
 
 4 
 
 2 
 
 0-9417 
 
 
 
 19 
 
 7 
 
 125 
 
 23 
 
 
 
 5 
 
 4 
 
 1 
 
 11-6143 
 
 
 
 19 
 
 6 
 
 500 
 
 23 
 
 
 
 4 
 
 4 
 
 1 
 
 10-2869 
 
 
 
 19 
 
 5 
 
 875 
 
 23 
 
 
 
 3 
 
 4 
 
 1 
 
 8-9595 
 
 
 
 19 
 
 5 
 
 250 
 
 23 
 
 
 
 2 
 
 4 
 
 1 
 
 7-6321 
 
 
 
 19 
 
 4 
 
 625 
 
 23 
 
 
 
 1 
 
 4 
 
 1 
 
 6-3047 
 
 
 
 19 
 
 4 
 
 000 
 
 23 
 
 
 
 
 
 4 
 
 1 
 
 4-9772 
 
 
 
 19 
 
 3 
 
 375 
 
 22 
 
 3 
 
 7 
 
 4 
 
 1 
 
 3-6498 
 
 
 
 19 
 
 2 
 
 750 
 
 22 
 
 3 
 
 6 
 
 4 
 
 1 
 
 2-3224 
 
GOLD-VALUING TABLE. 
 
 Ill 
 
 FINE GOLD, 
 
 Per Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 O*. 
 
 DivU 
 
 Grs. 
 
 Carats Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 19 
 
 2-125 
 
 22 
 
 3 
 
 5 
 
 4 
 
 1 
 
 0-9950 
 
 
 
 19 
 
 1-500 
 
 22 
 
 3 
 
 4 
 
 4 
 
 
 
 11-6676 
 
 
 
 19 
 
 0-875 
 
 22 
 
 3 
 
 3 
 
 4 
 
 
 
 10-3402 
 
 
 
 19 
 
 0-250 
 
 22 
 
 3 
 
 2 
 
 4 
 
 
 
 8-0127 
 
 
 
 18 
 
 23-625 
 
 22 
 
 3 
 
 1 
 
 4 
 
 
 
 7-6854 
 
 
 
 18 
 
 23-000 
 
 22 
 
 3 
 
 
 
 4 
 
 
 
 6-3579 
 
 
 
 18 
 
 22-375 
 
 22 
 
 2 
 
 7 
 
 4 
 
 
 
 4-0305 
 
 
 
 18 
 
 21-750 
 
 22 
 
 2 
 
 6 
 
 4 
 
 
 
 3-7031 
 
 
 
 18 
 
 21-125 
 
 22 
 
 2 
 
 5 
 
 4 
 
 
 
 2-3757 
 
 
 
 18 
 
 20-500 
 
 22 
 
 2 
 
 4 
 
 4 
 
 
 
 0-0482 
 
 
 
 18 
 
 19-875 
 
 22 
 
 2 
 
 3 
 
 3 
 
 19 
 
 11-7208 
 
 
 
 18 
 
 19-250 
 
 22 
 
 2 
 
 2 
 
 3 
 
 19 
 
 10-3934 
 
 
 
 18 
 
 18-625 
 
 22 
 
 2 
 
 1 
 
 3 
 
 19 
 
 8-0660 
 
 
 
 18 
 
 18-000 
 
 22 
 
 2 
 
 
 
 3 
 
 19 
 
 7-7386 
 
 
 
 18 
 
 17-375 
 
 22 
 
 1 
 
 7 
 
 3 
 
 19 
 
 6-4112 
 
 
 
 18 
 
 16-750 
 
 22 
 
 1 
 
 6 
 
 3 
 
 19 
 
 4-0838 
 
 
 
 18 
 
 16-125 
 
 22 
 
 1 
 
 5 
 
 3 
 
 19 
 
 3-7563 
 
 
 
 18 
 
 15-500 
 
 22 
 
 1 
 
 4 
 
 3 
 
 19 
 
 2-4289 
 
 
 
 18 
 
 14-875 
 
 22 
 
 1 
 
 3 
 
 3 
 
 19 
 
 0-1015 
 
 
 
 18 
 
 14-250 
 
 22 
 
 1 
 
 2 
 
 3 
 
 18 
 
 11-7741 
 
 
 
 18 
 
 13-625 
 
 22 
 
 1 
 
 1 
 
 3 
 
 18 
 
 10-4467 
 
 
 
 18 
 
 13-000 
 
 22 
 
 1 
 
 
 
 3 
 
 18 
 
 8-1193 
 
 
 
 18 
 
 12-375 
 
 22 
 
 
 
 7 
 
 3 
 
 18 
 
 7-7919 
 
 
 
 18 
 
 11-750 
 
 22 
 
 
 
 6 
 
 3 
 
 18 
 
 6-4644 
 
 
 
 18 
 
 11-125 
 
 22 
 
 
 
 5 
 
 3 
 
 18 
 
 4-1370 
 
 
 
 18 
 
 10-500 
 
 22 
 
 
 
 4 
 
 3 
 
 18 
 
 3-8096 
 
 
 
 18 
 
 9-875 
 
 22 
 
 
 
 3 
 
 3 
 
 18 
 
 2-4822 
 
 
 
 18 
 
 9-250 
 
 22 
 
 
 
 2 
 
 3 
 
 18 
 
 0-1548 
 
 
 
 18 
 
 8-625 
 
 22 
 
 
 
 1 
 
 3 
 
 17 
 
 11-8274 
 
 
 
 18 
 
 8-000 
 
 22 
 
 
 
 
 
 3 
 
 17 
 
 10-5000 
 
 
 
 18 
 
 7-375 
 
 21 
 
 3 
 
 7 
 
 3 
 
 17 
 
 .8-1725 
 
 
 
 18 
 
 6-750 
 
 21 
 
 3 
 
 6 
 
 3 
 
 17 
 
 7-8451 
 
 
 
 18 
 
 6-125 
 
 21 
 
 3 
 
 5 
 
 3 
 
 17 
 
 6-5177 
 
 
 
 18 
 
 5-500 
 
 21 
 
 3 
 
 4 
 
 3 
 
 17 
 
 4-1903 
 
 
 
 18 
 
 4-875 
 
 21 
 
 3 
 
 3 
 
 3 
 
 17 
 
 3-8629 
 
 
 
 18 
 
 4-250 
 
 21 
 
 3 
 
 2 
 
 3 
 
 17 
 
 2-5355 
 
 
 
 18 
 
 3-625 
 
 21 
 
 3 
 
 1 
 
 3 
 
 17 
 
 0-2081 
 
 
 
 18 
 
 3-000 
 
 21 
 
 3 
 
 
 
 3 
 
 16 
 
 11-8806 
 
 
 
 18 
 
 2-375 
 
 21 
 
 2 
 
 7 
 
 3 
 
 16 
 
 10-5532 
 
 
 
 18 
 
 1-750 
 
 21 
 
 2 
 
 6 
 
 3 
 
 16 
 
 8-2258 
 
 
 
 18 
 
 1-125 
 
 21 
 
 2 
 
 5 
 
 3 
 
 16 
 
 7-8984 
 
 
 
 18 
 
 0-500 
 
 21 
 
 2 
 
 4 
 
 3 
 
 16 
 
 6-5710 
 
 
 
 17 
 
 23-875 
 
 21 
 
 2 
 
 3 
 
 3 
 
 16 
 
 4-2436 
 
 
 
 17 
 
 23-250 
 
 21 
 
 2 
 
 2 
 
 3 
 
 16 
 
 3-9162 
 
 
 
 
 
 
 
 ?> M 2 
 
IV 
 
 GOLD-VALUING TABLE. 
 
 
 FINE 
 
 Per 
 
 GrOLD, 
 
 Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Dn-ts. Grs. 
 
 Carats 
 
 Grs. 
 
 Eighths 
 
 s. 
 
 d. 
 
 
 
 17 
 
 22-625 
 
 21 
 
 2 
 
 1 
 
 3 16 
 
 2-5887 
 
 
 
 17 
 
 22-000 
 
 21 
 
 2 
 
 
 
 3 16 
 
 1-2613 
 
 
 
 17 
 
 21-375 
 
 21 
 
 1 
 
 7 
 
 3 15 
 
 11-9339 
 
 
 
 17 
 
 20-750 
 
 21 
 
 1 
 
 6 
 
 3 15 
 
 10-6065 
 
 
 
 17 
 
 20-125 
 
 21 
 
 1 
 
 5 
 
 3 15 
 
 9-2791 
 
 
 
 17 
 
 '19-500 
 
 21 
 
 1 
 
 4 
 
 3 15 
 
 7-9517 
 
 
 
 17 
 
 18-875 
 
 21 
 
 1 
 
 3 
 
 3 15 
 
 6-6243 
 
 
 
 17 
 
 18-250 
 
 21 
 
 1 
 
 2 
 
 3 15 
 
 5-2968 
 
 
 
 17 
 
 17-625 
 
 21 
 
 1 
 
 1 
 
 3 15 
 
 3-9694 
 
 
 
 17 
 
 17-000 
 
 21 
 
 1 
 
 
 
 3 15 
 
 2-6420 
 
 
 
 17 
 
 16-375 
 
 21 
 
 
 
 7 
 
 3 15 
 
 1-3146 
 
 
 
 17 
 
 15-750 
 
 21 
 
 
 
 6 
 
 3 14 
 
 11-9872 
 
 
 
 17 
 
 15-125 
 
 21 
 
 
 
 5 
 
 3 14 
 
 10-6598 
 
 
 
 17 
 
 14-500 
 
 21 
 
 
 
 4 
 
 3 14 
 
 9-3324 
 
 
 
 17 
 
 13-875 
 
 21 
 
 
 
 3 
 
 3 14 
 
 8-0049 
 
 
 
 17 
 
 13-250 
 
 21 
 
 
 
 2 
 
 3 14 
 
 6-6775 
 
 
 
 17 
 
 12-625 
 
 21 
 
 
 
 1 
 
 3 14 
 
 5-3501 
 
 
 
 17 
 
 12-000 
 
 21 
 
 
 
 
 
 3 14 
 
 4-0227 
 
 
 
 17 
 
 11-375 
 
 20 
 
 3 
 
 7 
 
 3 14 
 
 2-6953 
 
 
 
 17 
 
 10-750 
 
 20 
 
 3 
 
 6 
 
 3 14 
 
 1-3678 
 
 
 
 17 
 
 10-125 
 
 20 
 
 3 
 
 5 
 
 3 14 
 
 0-0404 
 
 
 
 17 
 
 9-500 
 
 20 
 
 3 
 
 4 
 
 3 13 
 
 10-7130 
 
 
 
 17 
 
 8-875 
 
 20 
 
 3 
 
 3 
 
 3 13 
 
 9-3856 
 
 
 
 17 
 
 8-250 
 
 20 
 
 3 
 
 2 
 
 3 13 
 
 8-0582 
 
 
 
 17 
 
 7-6^5 
 
 20 
 
 3 
 
 1 
 
 3 13 
 
 6-7308 
 
 
 
 17 
 
 7-000 
 
 20 
 
 3 
 
 
 
 3 13 
 
 5-4034 
 
 
 
 17 
 
 6-375 
 
 20 
 
 2 
 
 7 
 
 3 13 
 
 4-0759 
 
 
 
 17 
 
 5-750 
 
 20 
 
 2 
 
 6 
 
 3 13 
 
 2-7485 
 
 
 
 17 
 
 5-125 
 
 20 
 
 2 
 
 5 
 
 3 13 
 
 1-4211 
 
 
 
 17 
 
 4-500 
 
 20 
 
 2 
 
 4 
 
 3 13 
 
 0-0937 
 
 
 
 17 
 
 3-875 
 
 20 
 
 2 
 
 3 
 
 3 12 
 
 10-7663 
 
 
 
 17 
 
 3-250 
 
 20 
 
 2 
 
 2 
 
 3 12 
 
 9-4389 
 
 
 
 17 
 
 2-625 
 
 20 
 
 2 
 
 1 
 
 3 12 
 
 8-1115 
 
 
 
 17 
 
 2-000 
 
 20 
 
 2 
 
 
 
 3 12 
 
 6-7840 
 
 
 
 17 
 
 1-375 
 
 20 
 
 1 
 
 hr 
 t 
 
 3 12 
 
 5-4566 
 
 
 
 17 
 
 0-750 
 
 20 
 
 1 
 
 6 
 
 3 12 
 
 4-1292 
 
 
 
 17 
 
 0-125 
 
 20 
 
 1 
 
 5 
 
 3 12 
 
 2-8018 
 
 
 
 16 
 
 23-500 
 
 20 
 
 1 
 
 4 
 
 3 12 
 
 1-4744 
 
 
 
 16 
 
 22-875 
 
 20 
 
 1 
 
 3 
 
 3 12 
 
 0-1470 
 
 
 
 16 
 
 22-250 
 
 20 
 
 1 
 
 2 
 
 3 11 
 
 10-8196 
 
 
 
 16 
 
 21-625 
 
 20 
 
 1 
 
 1 
 
 3 11 
 
 9-4921 
 
 
 
 16 
 
 21-000 
 
 20 
 
 1 
 
 
 
 3 11 
 
 8-1647 
 
 
 
 16 
 
 20-375 
 
 20 
 
 
 
 7 
 
 3 11 
 
 6-8373 
 
 
 
 16 
 
 19-750 
 
 20 
 
 
 
 6 
 
 3 11 
 
 5-5099 
 
GOLD-VALUING TABLE. 
 
 
 FINE 
 
 Per 
 
 GrOLD, 
 
 Ounce 
 
 CARAT G-OLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Dmts 
 
 Grs. 
 
 Carats Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 16 
 
 19-125 
 
 20 
 
 
 
 5 
 
 3 
 
 11 
 
 4-1825 
 
 
 
 16 
 
 18-500 
 
 20 
 
 
 
 4 
 
 3 
 
 11 
 
 2-8551 
 
 
 
 16 
 
 17-875 
 
 20 
 
 
 
 3 
 
 3 
 
 11 
 
 1-5277 
 
 
 
 16 
 
 17-250 
 
 20 
 
 
 
 2 
 
 3 
 
 11 
 
 0-2002 
 
 
 
 16 
 
 16-625 
 
 20 
 
 
 
 1 
 
 3 
 
 10 
 
 10-8728 
 
 
 
 16 
 
 16-000 
 
 20 
 
 
 
 
 
 3 
 
 10 
 
 9-5454 
 
 
 
 16 
 
 15-375 
 
 19 
 
 3 
 
 7 
 
 3 
 
 10 
 
 8-2180 
 
 
 
 16 
 
 14-750 
 
 19 
 
 3 
 
 6 
 
 3 
 
 10 
 
 6-8906 
 
 
 
 16 
 
 14-125 
 
 19 
 
 3 
 
 5 
 
 3 
 
 10 
 
 5-5632 
 
 
 
 16 
 
 13-500 
 
 19 
 
 3 
 
 4 
 
 3 
 
 10 
 
 4-2357 
 
 
 
 16 
 
 12-875 
 
 19 
 
 3 
 
 3 
 
 3 
 
 10 
 
 2-9083 
 
 
 
 16 
 
 12-250 
 
 19 
 
 3 
 
 2 
 
 3 
 
 10 
 
 1-5809 
 
 
 
 16 
 
 11-625 
 
 19 
 
 3 
 
 1 
 
 3 
 
 10 
 
 0-2534 
 
 
 
 16 
 
 11-000 
 
 19 
 
 3 
 
 
 
 3 
 
 9 
 
 10-9260 
 
 
 
 16 
 
 10-375 
 
 19 
 
 2 
 
 7 
 
 3 
 
 9 
 
 9-5986 
 
 
 
 16 
 
 9-750 
 
 19 
 
 2 
 
 6 
 
 3 
 
 9 
 
 8-2712 
 
 
 
 16 
 
 9-125 
 
 19 
 
 2 
 
 5 
 
 3 
 
 9 
 
 6-9437 
 
 
 
 16 
 
 8-500 
 
 19 
 
 2 
 
 4 
 
 3 
 
 9 
 
 5-6163 
 
 o 
 
 16 
 
 7-875 
 
 19 
 
 2 
 
 3 
 
 3 
 
 9 
 
 4-2889 
 
 
 
 16 
 
 7-250 
 
 19 
 
 2 
 
 2 
 
 3 
 
 9 
 
 2-9615 
 
 
 
 16 
 
 6-625 
 
 19 
 
 2 
 
 1 
 
 3 
 
 9 
 
 1-6341 
 
 
 
 16 
 
 6-000 
 
 19 
 
 2 
 
 
 
 3 
 
 9 
 
 0-3067 
 
 
 
 16 
 
 5-375 
 
 19 
 
 1 
 
 7 
 
 3 
 
 8 
 
 10-9793 
 
 
 
 16 
 
 4-750 
 
 19 
 
 1 
 
 6 
 
 3 
 
 8 
 
 9-6518 
 
 
 
 16 
 
 4-125 
 
 19 
 
 1 
 
 5 
 
 3 
 
 8 
 
 8-3244 
 
 
 
 16 
 
 3-500 
 
 19 
 
 1 
 
 4 
 
 3 
 
 8 
 
 6-9970 
 
 
 
 16 
 
 2-875 
 
 19 
 
 1 
 
 3 
 
 3 
 
 8 
 
 5-6696 
 
 
 
 16 
 
 2-250 
 
 19 
 
 1 
 
 2 
 
 3 
 
 8 
 
 4-3422 
 
 
 
 16 
 
 1-625 
 
 19 
 
 1 
 
 1 
 
 3 
 
 8 
 
 3-0148 
 
 
 
 16 
 
 1-000 
 
 19 
 
 1 
 
 
 
 3 
 
 8 
 
 1-6874 
 
 
 
 16 
 
 0-375 
 
 19 
 
 
 
 7 
 
 3 
 
 8 
 
 0-3599 
 
 
 
 15 
 
 23-750 
 
 19 
 
 
 
 6 
 
 3 
 
 7 
 
 11-0325 
 
 
 
 15 
 
 23-125 
 
 19 
 
 
 
 5 
 
 3 
 
 7 
 
 9-7051 
 
 
 
 15 
 
 22-500 
 
 19 
 
 
 
 4 
 
 3 
 
 7 
 
 8-3777 
 
 
 
 15 
 
 21-875 
 
 19 
 
 
 
 3 
 
 3 
 
 7 
 
 7-0503 
 
 
 
 15 
 
 21-250 
 
 19 
 
 
 
 2 
 
 3 
 
 7 
 
 5-7229 
 
 
 
 15 
 
 20-625 
 
 19 
 
 
 
 1 
 
 3 
 
 7 
 
 4-3955 
 
 
 
 15 
 
 20-000 
 
 19 
 
 
 
 
 
 3 
 
 7 
 
 3-0681 
 
 
 
 15 
 
 19-375 
 
 18 
 
 3 
 
 7 
 
 3 
 
 7 
 
 1-7407 
 
 
 
 15 
 
 18-750 
 
 18 
 
 3 
 
 6 
 
 3 
 
 7 
 
 0-4133 
 
 
 
 15 
 
 18-125 
 
 18 
 
 3 
 
 5 
 
 3 
 
 6 
 
 11-0859 
 
 
 
 15 
 
 17-500 
 
 18 
 
 3 
 
 4 
 
 3 
 
 6 
 
 9-7585 
 
 
 
 15 
 
 16-875 
 
 18 
 
 3 
 
 3 
 
 3 
 
 6 
 
 8-4311 
 
 
 
 15 
 
 16-250 
 
 18 
 
 3 
 
 2 
 
 3 
 
 6 
 
 7-1036 
 
VI 
 
 GOLD-VALUING TABLE. 
 
 
 FINE 
 
 GOLD, 
 
 CARAT GOLD, 
 
 STERLING VALUE, 
 
 
 Per Ounce 
 
 Per 
 
 Ounce 
 
 
 Per 
 
 Ounce 
 
 6b. 
 
 JJivts. 
 
 Grs. 
 
 Car a 's 
 
 Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 15 
 
 15-625 
 
 18 
 
 3 
 
 1 
 
 3 
 
 6 
 
 5-7762 
 
 
 
 15 
 
 15-000 
 
 18 
 
 3 
 
 
 
 3 
 
 6 
 
 4-4488 
 
 
 
 15 
 
 14-375 
 
 18 
 
 2 
 
 7 
 
 3 
 
 6 
 
 3-1214 
 
 
 
 15 
 
 13-750 
 
 18 
 
 2 
 
 6 
 
 3 
 
 6 
 
 1-7940 
 
 
 
 15 
 
 13-125 
 
 18 
 
 2 
 
 5 
 
 3 
 
 6 
 
 0-4666 
 
 
 
 15 
 
 12-500 
 
 18 
 
 2 
 
 4 
 
 3 
 
 5 
 
 11-1392 
 
 
 
 15 
 
 11-875 
 
 18 
 
 2 
 
 3 
 
 3 
 
 5 
 
 9-8117 
 
 
 
 15 
 
 11-250 
 
 18 
 
 2 
 
 2 
 
 3 
 
 5 
 
 8-4843 
 
 
 
 15 
 
 10-625 
 
 18 
 
 2 
 
 1 
 
 3 
 
 5 
 
 7-1569 
 
 
 
 15 
 
 10-000 
 
 18 
 
 2 
 
 
 
 3 
 
 5 
 
 5-8295 
 
 
 
 15 
 
 9-375 
 
 18 
 
 1 
 
 7 
 
 3 
 
 5 
 
 4-5021 
 
 
 
 15 
 
 8-750 
 
 18 
 
 1 
 
 6 
 
 3 
 
 5 
 
 3-1747 
 
 
 
 15 
 
 8-125 
 
 18 
 
 1 
 
 5 
 
 3 
 
 5 
 
 1-8473 
 
 
 
 15 
 
 7-500 
 
 18 
 
 1 
 
 4 
 
 3 
 
 5 
 
 0-5198 
 
 
 
 15 
 
 6-875 
 
 18 
 
 1 
 
 3 
 
 3 
 
 4 
 
 11-1924 
 
 
 
 15 
 
 6-250 
 
 18 
 
 1 
 
 2 
 
 3 
 
 4 
 
 9-8650 
 
 
 
 15 
 
 5-625 
 
 18 
 
 1 
 
 1 
 
 3 
 
 4 
 
 8-5376 
 
 
 
 15 
 
 5-000 
 
 18 
 
 1 
 
 
 
 3 
 
 4 
 
 7-2102 
 
 
 
 15 
 
 4-375 
 
 18 
 
 
 
 7 
 
 3 
 
 4 
 
 5-8828 
 
 
 
 15 
 
 3-750 
 
 18 
 
 
 
 6 
 
 3 
 
 4 
 
 4-5554 
 
 
 
 15 
 
 3-125 
 
 18 
 
 
 
 5 
 
 3 
 
 4 
 
 3-2279 
 
 
 
 15 
 
 2-500 
 
 18 
 
 
 
 4 
 
 3 
 
 4 
 
 1-9005 
 
 
 
 15 
 
 1-875 
 
 18 
 
 
 
 3 
 
 3 
 
 4 
 
 0-5731 
 
 
 
 15 
 
 1-250 
 
 18 
 
 
 
 2 
 
 3 
 
 3 
 
 11-2457 
 
 
 
 15 
 
 0-625 
 
 18 
 
 
 
 1 
 
 3 
 
 3 
 
 9-9183 
 
 
 
 15 
 
 o-ooo 
 
 18 
 
 
 
 
 
 3 
 
 3 
 
 8-5909 
 
 
 
 14 
 
 23-375 
 
 17 
 
 3 
 
 7 
 
 3 
 
 3 
 
 7-2634 
 
 
 
 14 
 
 22-750 
 
 17 
 
 3 
 
 6 
 
 3 
 
 3 
 
 5-9360 
 
 
 
 14 
 
 22-125 
 
 17 
 
 3 
 
 5 
 
 3 
 
 3 
 
 4-6086 
 
 
 
 14 
 
 21-500 
 
 17 
 
 3 
 
 4 
 
 3 
 
 3 
 
 3-2812 
 
 
 
 14 
 
 20-875 
 
 17 
 
 3 
 
 3 
 
 3 
 
 3 
 
 1-9538 
 
 
 
 14 
 
 20-250 
 
 17 
 
 3 
 
 2 
 
 3 
 
 3 
 
 0-6264 
 
 
 
 14 
 
 19-625 
 
 17 
 
 3 
 
 1 
 
 3 
 
 2 
 
 11-2990 
 
 
 
 14 
 
 19-000 
 
 17 
 
 3 
 
 
 
 3 
 
 2 
 
 9-9715 
 
 
 
 14 
 
 18-375 
 
 17 
 
 2 
 
 7 
 
 3 
 
 2 
 
 8-6441 
 
 
 
 14 
 
 17-750 
 
 17 
 
 2 
 
 6 
 
 3 
 
 2 
 
 7-3167 
 
 
 
 14 
 
 17-125 
 
 17 
 
 2 
 
 5 
 
 3 
 
 2 
 
 5-9893 
 
 
 
 14 
 
 16-500 
 
 17 
 
 2 
 
 4 
 
 3 
 
 2 
 
 4-6619 
 
 
 
 14 
 
 15-875 
 
 17 
 
 2 
 
 3 
 
 3 
 
 2 
 
 3-3345 
 
 
 
 14 
 
 15-250 
 
 17 
 
 2 
 
 2 
 
 3 
 
 2 
 
 2-0071 
 
 
 
 14 
 
 14-625 
 
 17 
 
 2 
 
 1 
 
 3 
 
 2 
 
 0-6796 
 
 
 
 14 
 
 14-000 
 
 17 
 
 2 
 
 
 
 3 
 
 1 
 
 11-3522 
 
 
 
 14 
 
 13-375 
 
 17 
 
 1 
 
 7 
 
 3 
 
 1 
 
 10-0248 
 
 
 
 14 
 
 12-750 
 
 17 
 
 1 
 
 6 
 
 3 
 
 1 
 
 8-6974 
 
GOLD-VALUING TABLE. 
 
 Vll 
 
 FINE G-OLD, 
 
 Per Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Divts. 
 
 Grs. 
 
 Carats Grs, 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 14 
 
 12-125 
 
 17 
 
 1 
 
 5 
 
 3 
 
 I 
 
 7-3700 
 
 
 
 14 
 
 11-500 
 
 17 
 
 1 
 
 4 
 
 3 
 
 1 
 
 6-0426 
 
 
 
 14 
 
 10-875 
 
 17 
 
 1 
 
 3 
 
 3 
 
 1 
 
 4-7152 
 
 
 
 14 
 
 10-250 
 
 17 
 
 1 
 
 2 
 
 3 
 
 1 
 
 3-3877 
 
 
 
 14 
 
 9-625 
 
 17 
 
 1 
 
 1 
 
 3 
 
 1 
 
 2-0603 
 
 
 
 14 
 
 9-000 
 
 17 
 
 1 
 
 
 
 3 
 
 1 
 
 0-7329 
 
 
 
 14 
 
 8-375 
 
 17 
 
 
 
 7 
 
 3 
 
 
 
 11-4055 
 
 
 
 14 
 
 7-750 
 
 17 
 
 
 
 6 
 
 3 
 
 
 
 10-0781 
 
 
 
 14 
 
 7-125 
 
 17 
 
 
 
 5 
 
 3 
 
 
 
 8-7507 
 
 
 
 14 
 
 6-500 
 
 17 
 
 
 
 4 
 
 3 
 
 
 
 7-4233 
 
 
 
 14 
 
 5-875 
 
 17 
 
 
 
 3 
 
 3 
 
 
 
 6-0958 
 
 
 
 14 
 
 5-250 
 
 17 
 
 
 
 2 
 
 3 
 
 
 
 4-7684 
 
 
 
 14 
 
 4-625 
 
 17 
 
 
 
 1 
 
 3 
 
 
 
 3-4410 
 
 
 
 14 
 
 4-000 
 
 17 
 
 
 
 
 
 3 
 
 
 
 2-1136 
 
 
 
 14 
 
 3-375 
 
 16 
 
 3 
 
 7 
 
 3 
 
 
 
 0-7862 
 
 
 
 14 
 
 2-750 
 
 16 
 
 3 
 
 6 
 
 2 
 
 19 
 
 11-4588 
 
 
 
 14 
 
 2-125 
 
 16 
 
 3 
 
 5 
 
 2 
 
 19 
 
 10-1313 
 
 
 
 14 
 
 1-500 
 
 16 
 
 3 
 
 4 
 
 2 
 
 19 
 
 8-8039 
 
 
 
 14 
 
 0-875 
 
 16 
 
 3 
 
 3 
 
 2 
 
 19 
 
 7-4765 
 
 
 
 14 
 
 0-250 
 
 16 
 
 3 
 
 2 
 
 2 
 
 19 
 
 6-1491 
 
 
 
 13 
 
 23-625 
 
 16 
 
 3 
 
 1 
 
 2 
 
 19 
 
 4-8217 
 
 
 
 13 
 
 23-000 
 
 16 
 
 3 
 
 
 
 2 
 
 19 
 
 3-4943 
 
 
 
 13 
 
 22-375 
 
 16 
 
 2 
 
 7 
 
 2 
 
 19 
 
 2-1669 
 
 
 
 13 
 
 21-750 
 
 16 
 
 2 
 
 6 
 
 2 
 
 19 
 
 0-8394 
 
 
 
 13 
 
 21-125 
 
 16 
 
 2 
 
 5 
 
 2 
 
 18 
 
 11-5120 
 
 
 
 13 
 
 20-500 
 
 16 
 
 2 
 
 4 
 
 2 
 
 18 
 
 10-1846 
 
 
 
 13 
 
 19-875 
 
 16 
 
 2 
 
 3 
 
 2 
 
 18 
 
 8-8572 
 
 
 
 13 
 
 19-250 
 
 16 
 
 2 
 
 2 
 
 2 
 
 18 
 
 7-5298 
 
 
 
 13 
 
 18-625 
 
 16 
 
 2 
 
 1 
 
 2 
 
 18 
 
 6-2024 
 
 
 
 13 
 
 18-000 
 
 16 
 
 2 
 
 
 
 2 
 
 18 
 
 4-8750 
 
 
 
 13 
 
 17-375 
 
 16 
 
 1 
 
 7 
 
 2 
 
 18 
 
 3-5475 
 
 
 
 13 
 
 16-750 
 
 16 
 
 1 
 
 6 
 
 2 
 
 18 
 
 2-2201 
 
 
 
 13 
 
 16-125 
 
 16 
 
 1 
 
 5 
 
 2 
 
 18 
 
 08927 
 
 
 
 13 
 
 15-500 
 
 16 
 
 1 
 
 4 
 
 2 
 
 17 
 
 11-5653 
 
 
 
 13 
 
 14-875 
 
 16 
 
 1 
 
 3 
 
 2 
 
 17 
 
 10-2377 
 
 
 
 13 
 
 14-250 
 
 16 
 
 1 
 
 2 
 
 2 
 
 17 
 
 8-9103 
 
 
 
 13 
 
 13-625 
 
 16 
 
 1 
 
 1 
 
 2 
 
 17 
 
 7-5829 
 
 
 
 13 
 
 13-000 
 
 16 
 
 1 
 
 
 
 2 
 
 17 
 
 6-2554 
 
 
 
 13 
 
 12-375 
 
 16 
 
 
 
 7 
 
 2 
 
 17 
 
 4-9280 
 
 
 
 13 
 
 11-750 
 
 16 
 
 
 
 6 
 
 2 
 
 17 
 
 3-6006 
 
 
 
 13 
 
 11-125 
 
 16 
 
 
 
 5 
 
 2 
 
 17 
 
 2-2732 
 
 
 
 13 
 
 10-500 
 
 16 
 
 
 
 4 
 
 2 
 
 17 
 
 0-9458 
 
 
 
 13 
 
 9-875 
 
 16 
 
 
 
 3 
 
 2 
 
 16 
 
 11-6184 
 
 
 
 13 
 
 9-250 16 
 
 
 
 2 
 
 2 
 
 16 
 
 10-2909 
 
Vlll 
 
 GOLD-VALUING TABLE. 
 
 
 FINE 
 Per 
 
 GOLD, 
 
 Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Divts. 6frs. 
 
 Carats Grs. M 
 
 \ghtJis 
 
 
 
 s. 
 
 d. 
 
 
 
 13 
 
 8-625 
 
 16 
 
 
 
 1 
 
 2 
 
 16 
 
 8-9635 
 
 
 
 13 
 
 8-000 
 
 16 
 
 
 
 
 
 2 
 
 16 
 
 7-6363 
 
 
 
 13 
 
 7-375 
 
 15 
 
 3 
 
 7 
 
 2 
 
 16 
 
 6-3089 
 
 
 
 13 
 
 6-750 
 
 15 
 
 3 
 
 6 
 
 2 
 
 16 
 
 4-9815 
 
 
 
 13 
 
 6-125 
 
 15 
 
 3 
 
 5 
 
 2 
 
 16 
 
 3-6541 
 
 
 
 13 
 
 5-500 
 
 15 
 
 3 
 
 4 
 
 2 
 
 16 
 
 2-3267 
 
 
 
 13 
 
 4-875 
 
 15 
 
 3 
 
 3 
 
 2 
 
 16 
 
 0-9992 
 
 
 
 13 
 
 4-250 
 
 15 
 
 3 
 
 2 
 
 2 
 
 15 
 
 11-6718 
 
 
 
 13 
 
 3-625 
 
 15 
 
 3 
 
 1 
 
 2 
 
 15 
 
 10-3444 
 
 
 
 13 
 
 3-000 
 
 15 
 
 3 
 
 
 
 2 
 
 15 
 
 9-0170 
 
 
 
 13 
 
 2-373 
 
 15 
 
 2 
 
 7 
 
 2 
 
 15 
 
 7-6896 
 
 
 
 13 
 
 1-750 
 
 15 
 
 2 
 
 6 
 
 2 
 
 15 
 
 6-3622 
 
 
 
 13 
 
 1-125 
 
 15 
 
 2 
 
 5 
 
 2 
 
 15 
 
 5-0348 
 
 
 
 13 
 
 0-500 
 
 15 
 
 2 
 
 4 
 
 2 
 
 15 
 
 3-7073 
 
 
 
 12 
 
 23-875 
 
 15 
 
 2 
 
 3 
 
 2 
 
 15 
 
 2-3799 
 
 
 
 12 
 
 23-250 
 
 15 
 
 2 
 
 2 
 
 2 
 
 15 
 
 1-0525 
 
 
 
 12 
 
 22-625 
 
 15 
 
 2 
 
 1 
 
 2 
 
 14 
 
 11-7251 
 
 
 
 12 
 
 22-000 
 
 15 
 
 2 
 
 
 
 2 
 
 14 
 
 10-3976 
 
 
 
 12 
 
 21-375 
 
 15 
 
 1 
 
 7 
 
 2 
 
 14 
 
 9-0702 
 
 
 
 12 
 
 20-750 
 
 15 
 
 1 
 
 6 
 
 2 
 
 14 
 
 7-7428 
 
 
 
 12 
 
 20-125 
 
 15 
 
 1 
 
 5 
 
 2 
 
 14 
 
 6-4154 
 
 
 
 12 
 
 19-500 
 
 15 
 
 1 
 
 4 
 
 2 
 
 14 
 
 5-0880 
 
 
 
 12 
 
 18-875 
 
 15 
 
 1 
 
 3 
 
 2 
 
 14 
 
 3-7606 
 
 
 
 12 
 
 18-250 
 
 15 
 
 1 
 
 2 
 
 2 
 
 14 
 
 2-4332 
 
 
 
 12 
 
 17-625 
 
 15 
 
 1 
 
 1 
 
 2 
 
 14 
 
 1-1057 
 
 
 
 12 
 
 17-000 
 
 15 
 
 1 
 
 
 
 2 
 
 13 
 
 11-7783 
 
 
 
 12 
 
 16-375 
 
 15 
 
 
 
 if 
 i 
 
 2 
 
 13 
 
 10-4509 
 
 
 
 12 
 
 15-750 
 
 15 
 
 
 
 6 
 
 2 
 
 13 
 
 9-1235 
 
 
 
 12 
 
 15-125 
 
 15 
 
 
 
 5 
 
 2 
 
 13 
 
 7-7961 
 
 
 
 12 
 
 14-500 
 
 15 
 
 
 
 4 
 
 2 
 
 13 
 
 6-4687 
 
 
 
 12 
 
 13-875 
 
 15 
 
 
 
 3 
 
 2 
 
 13 
 
 5-1413 
 
 
 
 12 
 
 13-250 
 
 15 
 
 
 
 2 
 
 2 
 
 13 
 
 3-8138 
 
 
 
 12 
 
 12-625 
 
 15 
 
 
 
 1 
 
 2 
 
 13 
 
 2-4864 
 
 
 
 12 
 
 12-000 
 
 15 
 
 
 
 
 
 2 
 
 13 
 
 1-1591 
 
 
 
 12 
 
 11-375 
 
 14 
 
 3 
 
 7 
 
 2 
 
 12 
 
 11-8316 
 
 
 
 12 
 
 10-750 
 
 14 
 
 3 
 
 6 
 
 2 
 
 12 
 
 10-5042 
 
 
 
 12 
 
 10-125 
 
 14 
 
 3 
 
 5 
 
 2 
 
 12 
 
 9-1768 
 
 
 
 12 
 
 9-500 
 
 14 
 
 3 
 
 4 
 
 2 
 
 12 
 
 7-8494 
 
 
 
 12 
 
 8-875 
 
 14 
 
 3 
 
 3 
 
 2 
 
 12 
 
 6-5220 
 
 
 
 12 
 
 8-250 
 
 14 
 
 3 
 
 2 
 
 2 
 
 12 
 
 5-1946 
 
 
 
 12 
 
 7-625 
 
 14 
 
 3 
 
 1 
 
 2 
 
 12 
 
 3-8671 
 
 
 
 12 
 
 7-000 
 
 14 
 
 3 
 
 
 
 2 
 
 12 
 
 2-5397 
 
 
 
 12 
 
 6-375 
 
 14 
 
 2 
 
 7 
 
 2 
 
 12 
 
 1-2123 
 
 
 
 12 
 
 5-750 
 
 14 
 
 2 
 
 6 
 
 2 
 
 11 
 
 11-8849 
 
GOLD-VALUING TABLE. 
 
 
 
 FINE GOLD, 
 
 Per Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 Carats Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 12 
 
 5-125 
 
 14 
 
 2 
 
 5 
 
 2 
 
 11 
 
 10-5575 
 
 
 
 12 
 
 4-500 
 
 14 
 
 2 
 
 4 
 
 2 
 
 11 
 
 9-2301 
 
 
 
 12 
 
 3-875 
 
 14 
 
 2 
 
 3 
 
 2 
 
 11 
 
 7-9027 
 
 
 
 12 
 
 3-250 
 
 14 
 
 2 
 
 2 
 
 2 
 
 11 
 
 6-5752 
 
 
 
 12 
 
 2-625 
 
 14 
 
 2 
 
 1 
 
 2 
 
 11 
 
 5-2478 
 
 
 
 12 
 
 2-000 
 
 14 
 
 2 
 
 
 
 2 
 
 11 
 
 3-9204 
 
 
 
 12 
 
 1-375 
 
 14 
 
 1 
 
 7 
 
 2 
 
 11 
 
 2-5930 
 
 
 
 12 
 
 0-750 
 
 14 
 
 1 
 
 6 
 
 2 
 
 11 
 
 1-2656 
 
 
 
 12 
 
 0-125 
 
 14 
 
 1 
 
 5 
 
 2 
 
 10 
 
 11-9382 
 
 
 
 11 
 
 23-500 
 
 14 
 
 1 
 
 4 
 
 2 
 
 10 
 
 10-6107 
 
 
 
 11 
 
 22-875 
 
 14 
 
 1 
 
 3 
 
 2 
 
 10 
 
 9-2833 
 
 
 
 11 
 
 22-250 
 
 14 
 
 1 
 
 2 
 
 '2 
 
 10 
 
 7-9559 
 
 
 
 11 
 
 21-625 
 
 14 
 
 1 
 
 1 
 
 2 
 
 10 
 
 6-6285 
 
 
 
 11 
 
 21-000 
 
 14 
 
 1 
 
 
 
 2 
 
 10 
 
 5-3011 
 
 
 
 11 
 
 20-375 
 
 14 
 
 
 
 7 
 
 2 
 
 10 
 
 3-9737 
 
 
 
 11 
 
 19-750 
 
 14 
 
 
 
 6 
 
 2 
 
 10 
 
 2-6463 
 
 
 
 11 
 
 19-125 
 
 14 
 
 
 
 5 
 
 2 
 
 10 
 
 1-3188 
 
 
 
 11 
 
 18-500 
 
 14 
 
 
 
 4 
 
 2 
 
 9 
 
 11-9914 
 
 
 
 11 
 
 17-875 
 
 14 
 
 
 
 3 
 
 2 
 
 9 
 
 10-6640 
 
 
 
 11 
 
 17-250 
 
 14 
 
 
 
 2 
 
 2 
 
 9 
 
 9-3366 
 
 
 
 11 
 
 16-625 
 
 14 
 
 
 
 1 
 
 2 
 
 9 
 
 8-0092 
 
 
 
 11 
 
 16-000 
 
 14 
 
 
 
 
 
 2 
 
 9 
 
 6-6818 
 
 
 
 11 
 
 15-375 
 
 13 
 
 3 
 
 7 
 
 2 
 
 9 
 
 5-3544 
 
 
 
 11 
 
 14-750 
 
 13 
 
 3 
 
 6 
 
 2 
 
 9 
 
 4-0269 
 
 
 
 11 
 
 14-150 
 
 13 
 
 3 
 
 5 
 
 2 
 
 9 
 
 2-6995 
 
 
 
 11 
 
 13-500 
 
 13 
 
 3 
 
 4 
 
 2 
 
 9 
 
 1-3721 
 
 
 
 11 
 
 12-875 
 
 13 
 
 3 
 
 3 
 
 2 
 
 9 
 
 0-0447 
 
 
 
 11 
 
 12-250 
 
 13 
 
 3 
 
 2 
 
 2 
 
 8 
 
 10-7173 
 
 
 
 11 
 
 11-625 
 
 13 
 
 3 
 
 1 
 
 2 
 
 8 
 
 9-3899 
 
 
 
 11 
 
 11-000 
 
 13 
 
 3 
 
 
 
 2 
 
 8 
 
 8-0625 
 
 
 
 11 
 
 10-375 
 
 13 
 
 2 
 
 7 
 
 2 
 
 8 
 
 6-7350 
 
 
 
 11 
 
 9-750 
 
 13 
 
 2 
 
 6 
 
 2 
 
 8 
 
 5-4076 
 
 
 
 11 
 
 9-125 
 
 13 
 
 2 
 
 5 
 
 2 
 
 8 
 
 4-0802 
 
 
 
 11 
 
 8-500 
 
 13 
 
 2 
 
 4 
 
 2 
 
 8 
 
 2-7528 
 
 
 
 11 
 
 7-875 
 
 13 
 
 2 
 
 3 
 
 2 
 
 8 
 
 1-4254 
 
 
 
 11 
 
 7-250 
 
 13 
 
 2 
 
 2 
 
 2 
 
 8 
 
 0-0980 
 
 
 
 11 
 
 6-625 
 
 13 
 
 2 
 
 1 
 
 2 
 
 7 
 
 10-7705 
 
 
 
 11 
 
 6-000 
 
 13 
 
 2 
 
 
 
 2 
 
 7 
 
 9-4431 
 
 
 
 11 
 
 5-375 
 
 13 
 
 1 
 
 7 
 
 2 
 
 7 
 
 8-1157 
 
 
 
 11 
 
 4-750 
 
 13 
 
 1 
 
 6 
 
 2 
 
 7 
 
 6-7883 
 
 
 
 11 
 
 4-125 
 
 13 
 
 1 
 
 5 
 
 2 
 
 7 
 
 5-4609 
 
 
 
 11 
 
 3-500 
 
 13 
 
 1 
 
 4 
 
 2 
 
 7 
 
 4-1335 
 
 
 
 11 
 
 2-875 
 
 13 
 
 1 
 
 3 
 
 2 
 
 7 
 
 2-8061 
 
 
 
 11 
 
 2-250 
 
 13 
 
 1 
 
 2 
 
 2 
 
 7 
 
 1-4786 
 
GOLD-VALUING TABLE. 
 
 FINE GOLD, 
 
 Per Ounce 
 
 CAKAT GOLD, 
 
 Per Ounce 
 
 STERLING VAI/~E, 
 
 Per Our-* 
 
 Oz. 
 
 Drvts. 
 
 Grs. 
 
 Carats Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 11 
 
 1-625 
 
 13 
 
 1 
 
 1 
 
 2 
 
 7 
 
 0-1512 
 
 
 
 11 
 
 1-000 
 
 13 
 
 1 
 
 
 
 2 
 
 6 
 
 10-8238 
 
 
 
 11 
 
 0-375 
 
 13 
 
 
 
 7 
 
 2 
 
 6 
 
 9-4964 
 
 
 
 10 
 
 23-750 
 
 13 
 
 
 
 6 
 
 2 
 
 6 
 
 8-1698 
 
 
 
 10 
 
 23-125 
 
 13 
 
 
 
 5 
 
 2 
 
 6 
 
 6-8416 
 
 
 
 10 
 
 22-500 
 
 13 
 
 
 
 4 
 
 2 
 
 6 
 
 5-5142 
 
 
 
 10 
 
 21-875 
 
 13 
 
 
 
 3 
 
 2 
 
 6 
 
 4-1867 
 
 
 
 10 
 
 21-250 
 
 13 
 
 
 
 2 
 
 2 
 
 6 
 
 2-8593 
 
 
 
 10 
 
 20-625 
 
 13 
 
 
 
 1 
 
 2 
 
 6 
 
 1-5319 
 
 
 
 10 
 
 20-000 
 
 13 
 
 
 
 
 
 2 
 
 6 
 
 0-2045 
 
 
 
 10 
 
 19-375 
 
 12 
 
 3 
 
 7 
 
 2 
 
 5 
 
 10-8771 
 
 
 
 10 
 
 18-750 
 
 12 
 
 3 
 
 6 
 
 2 
 
 5 
 
 9-5497 
 
 
 
 10 
 
 18-125 
 
 12 
 
 3 
 
 5 
 
 2 
 
 5 
 
 8-2223 
 
 
 
 10 
 
 17-500 
 
 12 
 
 3 
 
 4 
 
 2 
 
 5 
 
 6-8948 
 
 
 
 10 
 
 16-875 
 
 12 
 
 3 
 
 3 
 
 2 
 
 5 
 
 5-5674 
 
 
 
 10 
 
 16-250 
 
 12 
 
 3 
 
 2 
 
 2 
 
 5 
 
 4-2400 
 
 
 
 10 
 
 15-625 
 
 12 
 
 3 
 
 1 
 
 2 
 
 5 
 
 2-9126 
 
 
 
 10 
 
 15-000 
 
 12 
 
 3 
 
 
 
 2 
 
 5 
 
 1-5852 
 
 
 
 10 
 
 14-375 
 
 12 
 
 2 
 
 7 
 
 2 
 
 5 
 
 0-2578 
 
 
 
 10 
 
 13-750 
 
 12 
 
 2 
 
 6 
 
 2 
 
 4 
 
 10-9303 
 
 
 
 10 
 
 13-125 
 
 12 
 
 2 
 
 5 
 
 2 
 
 4 
 
 9-6029 
 
 
 
 10 
 
 12-500 
 
 12 
 
 2 
 
 4 
 
 2 
 
 4 
 
 8-2755 
 
 
 
 10 
 
 11-875 
 
 12 
 
 2 
 
 3 
 
 2 
 
 4 
 
 6-9481 
 
 
 
 10 
 
 11-250 
 
 12 
 
 2 
 
 2 
 
 2 
 
 4 
 
 5-6207 
 
 
 
 10 
 
 10-625 
 
 12 
 
 2 
 
 1 
 
 2 
 
 4 
 
 4-2933 
 
 
 
 10 
 
 10-000 
 
 12 
 
 2 
 
 
 
 2 
 
 4 
 
 2-9659 
 
 
 
 10 
 
 9-375 
 
 12 
 
 1 
 
 7 
 
 2 
 
 4 
 
 1-6384 
 
 
 
 10 
 
 8-750 
 
 12 
 
 1 
 
 6 
 
 2 
 
 4 
 
 0-3110 
 
 
 
 10 
 
 8-125 
 
 12 
 
 1 
 
 5 
 
 2 
 
 3 
 
 10-8366 
 
 
 
 10 
 
 7-500 
 
 12 
 
 1 
 
 4 
 
 2 
 
 3 
 
 9-6562 
 
 
 
 10 
 
 6-875 
 
 12 
 
 1 
 
 3 
 
 2 
 
 3 
 
 8-3288 
 
 
 
 10 
 
 6-250 
 
 12 
 
 1 
 
 2 
 
 2 
 
 3 
 
 7-0014 
 
 
 
 10 
 
 5-625 
 
 12 
 
 1 
 
 1 
 
 2 
 
 3 
 
 5-6740 
 
 
 
 10 
 
 5-000 
 
 12 
 
 1 
 
 
 
 2 
 
 3 
 
 4-3465 
 
 
 
 10 
 
 4-375 
 
 12 
 
 
 
 7 
 
 2 
 
 3 
 
 3-0191 
 
 
 
 10 
 
 3-750 
 
 12 
 
 
 
 6 
 
 2 
 
 3 
 
 1-6917 
 
 
 
 10 
 
 3-125 
 
 12 
 
 
 
 5 
 
 2 
 
 3 
 
 0-3643 
 
 
 
 10 
 
 2-500 
 
 12 
 
 
 
 4 
 
 2 
 
 2 
 
 11-0369 
 
 
 
 10 
 
 1-875 
 
 12 
 
 
 
 3 
 
 2 
 
 2 
 
 9-7095 
 
 
 
 10 
 
 1-250 
 
 12 
 
 
 
 2 
 
 2 
 
 2 
 
 8-3821 
 
 
 
 10 
 
 0-625 
 
 12 
 
 
 
 1 
 
 2 
 
 2 
 
 7-0546 
 
 
 
 10 
 
 0-000 
 
 12 
 
 
 
 
 
 2 
 
 2 
 
 5-7272 
 
 
 
 9 
 
 23-375 
 
 11 
 
 3 
 
 7 
 
 2 
 
 2 
 
 4-3998 
 
 
 
 9 
 
 22-750 
 
 11 
 
 3 
 
 6 
 
 2 
 
 2 
 
 3-0724 
 
GOLD-VALUING TABLE. 
 
 FINE GOLD, 
 
 Per Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Drvts 
 
 Grs. 
 
 Carats 
 
 Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 9 
 
 22-125 
 
 11 
 
 3 
 
 5 
 
 2 
 
 2 
 
 1-7450 
 
 
 
 9 
 
 21-500 
 
 11 
 
 3 
 
 4 
 
 2 
 
 2 
 
 0-4176 
 
 
 
 9 
 
 21-875 
 
 11 
 
 3 
 
 3 
 
 2 
 
 1 
 
 11-0901 
 
 
 
 9 
 
 20-250 
 
 11 
 
 3 
 
 2 
 
 2 
 
 1 
 
 9-7627 
 
 
 
 9 
 
 19-625 
 
 11 
 
 3 
 
 1 
 
 2 
 
 1 
 
 8-4353 
 
 
 
 9 
 
 19-000 
 
 11 
 
 3 
 
 
 
 2 
 
 1 
 
 7-1079 
 
 
 
 9 
 
 18-375 
 
 11 
 
 2 
 
 7 
 
 2 
 
 1 
 
 5-7805 
 
 
 
 9 
 
 17-750 
 
 11 
 
 2 
 
 6 
 
 2 
 
 1 
 
 4-4531 
 
 
 
 9 
 
 17-125 
 
 11 
 
 2 
 
 5 
 
 2 
 
 1 
 
 3-1257 
 
 
 
 9 
 
 16-500 
 
 11 
 
 2 
 
 4 
 
 2 
 
 1 
 
 1-7982 
 
 
 
 9 
 
 15-875 
 
 11 
 
 2 
 
 3 
 
 2 
 
 1 
 
 0-4708 
 
 
 
 9 
 
 15-250 
 
 11 
 
 2 
 
 2 
 
 2 
 
 
 
 11-1434 
 
 
 
 9 
 
 14-625 
 
 11 
 
 2 
 
 1 
 
 2 
 
 
 
 9-8160 
 
 
 
 9 
 
 14-000 
 
 11 
 
 2 
 
 
 
 2 
 
 
 
 8-4886 
 
 
 
 9 
 
 13-375 
 
 11 
 
 1 
 
 7 
 
 2 
 
 
 
 7-1612 
 
 
 
 9 
 
 12-750 
 
 11 
 
 1 
 
 6 
 
 2 
 
 
 
 5-8338 
 
 
 
 9 
 
 12-125 
 
 11 
 
 1 
 
 5 
 
 2 
 
 
 
 4-5063 
 
 
 
 9 
 
 11-500 
 
 11 
 
 1 
 
 4 
 
 2 
 
 
 
 3-1789 
 
 
 
 9 
 
 10-875 
 
 11 
 
 1 
 
 3 
 
 2 
 
 
 
 1-8515 
 
 
 
 9 
 
 10-250 
 
 11 
 
 1 
 
 2 
 
 2 
 
 
 
 0-5241 
 
 
 
 9 
 
 9-625 
 
 11 
 
 1 
 
 1 
 
 1 
 
 19 
 
 11-1967 
 
 
 
 9 
 
 9-000 
 
 11 
 
 1 
 
 
 
 1 
 
 19 
 
 9-8693 
 
 
 
 9 
 
 8-375 
 
 11 
 
 
 
 . 7 
 
 1 
 
 19 
 
 8-5419 
 
 
 
 9- 
 
 7-750 
 
 11 
 
 
 
 6 
 
 1 
 
 19 
 
 7-2144 
 
 
 
 9 
 
 7-125 
 
 11 
 
 
 
 5 
 
 1 
 
 19 
 
 5-8870 
 
 
 
 9 
 
 6-500 
 
 11 
 
 
 
 4 
 
 1 
 
 19 
 
 4-5596 
 
 
 
 9 
 
 5-875 
 
 11 
 
 
 
 3 
 
 1 
 
 19 
 
 3-2322 
 
 
 
 9 
 
 5-250 
 
 11 
 
 
 
 2 
 
 1 
 
 19 
 
 1-9048 
 
 
 
 9 
 
 4-625 
 
 11 
 
 
 
 1 
 
 1 
 
 19 
 
 0-5774 
 
 
 
 9 
 
 4-000 
 
 11 
 
 
 
 
 
 1 
 
 18 
 
 11-2500 
 
 
 
 9 
 
 3-375 
 
 10 
 
 3 
 
 7 
 
 1 
 
 18 
 
 9-9225 
 
 
 
 9 
 
 2-750 
 
 10 
 
 3 
 
 6 
 
 1 
 
 18 
 
 8-5951 
 
 
 
 9 
 
 2-125 
 
 10 
 
 3 
 
 5 
 
 1 
 
 18 
 
 7-2677 
 
 
 
 9 
 
 1-500 
 
 10 
 
 3 
 
 4 
 
 1 
 
 18 
 
 5-9403 
 
 
 
 9 
 
 0-875 
 
 10 
 
 3 
 
 3 
 
 1 
 
 18 
 
 4-6129 
 
 
 
 9 
 
 0-250 
 
 10 
 
 3 
 
 2 
 
 1 
 
 18 
 
 3-2855 
 
 
 
 8 
 
 23-625 
 
 10 
 
 3 
 
 1 
 
 1 
 
 18 
 
 1-9580 
 
 
 
 8 
 
 23-000 
 
 10 
 
 3 
 
 
 
 1 
 
 18 
 
 0-6306 
 
 
 
 8 
 
 22-375 
 
 10 
 
 2 
 
 7 
 
 1 
 
 17 
 
 11-3032 
 
 
 
 8 
 
 21-750 
 
 10 
 
 2 
 
 6 
 
 1 
 
 17 
 
 9-9758 
 
 
 
 8 
 
 21-125 
 
 10 
 
 2 
 
 5 
 
 1 
 
 17 
 
 8-6484 
 
 
 
 8 
 
 20-500 
 
 10 
 
 2 
 
 4 
 
 1 
 
 17 
 
 7-3210 
 
 
 
 8 
 
 19-875 
 
 10 
 
 2 
 
 3 
 
 1 
 
 17 
 
 5-9936 
 
 
 
 8 
 
 19-250 
 
 10 
 
 2 
 
 2 
 
 1 
 
 17 
 
 4-6661 
 
Xll 
 
 GOLD-VALUING TABLE. 
 
 FINE GOLD, 
 
 Per Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Diets 
 
 . Grs. 
 
 Carats 
 
 Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 8 
 
 18-625 
 
 10 
 
 2 
 
 1 
 
 1 
 
 17 
 
 3-3387 
 
 
 
 8 
 
 18-000 
 
 10 
 
 2 
 
 
 
 1 
 
 17 
 
 2-0113 
 
 
 
 8 
 
 17-375 
 
 10 
 
 1 
 
 7 
 
 1 
 
 17 
 
 0-6839 
 
 
 
 8 
 
 16-750 
 
 10 
 
 1 
 
 6 
 
 1 
 
 16 
 
 11-3565 
 
 
 
 8 
 
 16-125 
 
 10 
 
 1 
 
 5 
 
 1 
 
 16 
 
 10-0291 
 
 
 
 8 
 
 15-500 
 
 10 
 
 1 
 
 4 
 
 1 
 
 16 
 
 8-7017 
 
 
 
 8 
 
 14-875 
 
 10 
 
 1 
 
 3 
 
 1 
 
 16 
 
 7-3742 
 
 
 
 8 
 
 14-250 
 
 10 
 
 1 
 
 2 
 
 1 
 
 16 
 
 6-0468 
 
 
 
 8 
 
 13-625 
 
 10 
 
 1 
 
 1 
 
 1 
 
 16 
 
 4-7194 
 
 
 
 8 
 
 13-000 
 
 10 
 
 1 
 
 
 
 1 
 
 16 
 
 3-3920 
 
 
 
 8 
 
 12-375 
 
 10 
 
 
 
 7 
 
 1 
 
 16 
 
 2-0646 
 
 
 
 8 
 
 11-750 
 
 10 
 
 
 
 6 
 
 1 
 
 16 
 
 0-7372 
 
 
 
 8 
 
 11-125 
 
 10 
 
 
 
 5 
 
 1 
 
 15 
 
 11-4098 
 
 
 
 8 
 
 10-500 
 
 10 
 
 
 
 4 
 
 1 
 
 15 
 
 10-0823 
 
 
 
 8 
 
 9-875 
 
 10 
 
 
 
 3 
 
 1 
 
 15 
 
 8-7549 
 
 
 
 8 
 
 9-250 
 
 10 
 
 
 
 2 
 
 1 
 
 15 
 
 7-4275 
 
 
 
 8 
 
 8-625 
 
 10 
 
 
 
 1 
 
 1 
 
 15 
 
 6-1001 
 
 
 
 8 
 
 8-000 
 
 10 
 
 
 
 
 
 1 
 
 15 
 
 4-7728 
 
 
 
 8 
 
 7-375 
 
 9 
 
 3 
 
 7 
 
 1 
 
 15 
 
 3-4454 
 
 
 
 8 
 
 6-750 
 
 9 
 
 3 
 
 6 
 
 1 
 
 15 
 
 2-1179 
 
 
 
 8 
 
 6-125 
 
 9 
 
 3 
 
 5 
 
 1 
 
 15 
 
 0-7905 
 
 
 
 8 
 
 5-500 
 
 9 
 
 3 
 
 4 
 
 1 
 
 14 
 
 11-4631 
 
 
 
 8 
 
 4-875 
 
 9 
 
 3 
 
 3 
 
 1 
 
 14 
 
 10-1357 
 
 
 
 8 
 
 4-250 
 
 9 
 
 3 
 
 2 
 
 1 
 
 14 
 
 8-8083 
 
 
 
 8 
 
 3-625 
 
 9 
 
 3 
 
 1 
 
 1 
 
 14 
 
 7-4809 
 
 
 
 8 
 
 3-000 
 
 9 
 
 3 
 
 
 
 1 
 
 14 
 
 6-1535 
 
 
 
 8 
 
 2-375 
 
 9 
 
 2 
 
 7 
 
 1 
 
 14 
 
 4-8260 . 
 
 
 
 8 
 
 1-750 
 
 9 
 
 2 
 
 6 
 
 1 
 
 14 
 
 3-4986 
 
 
 
 8 
 
 1-125 
 
 9 
 
 2 
 
 5 
 
 1 
 
 14 
 
 2-1712 
 
 
 
 8 
 
 0-500 
 
 9 
 
 2 
 
 4 
 
 1 
 
 14 
 
 0-8438 
 
 
 
 7 
 
 23-875 
 
 9 
 
 2 
 
 3 
 
 1 
 
 13 
 
 11-5164 
 
 
 
 7 
 
 23-250 
 
 9 
 
 2 
 
 2 
 
 1 
 
 13 
 
 10-1890 
 
 
 
 7 
 
 22-625 
 
 9 
 
 2 
 
 1 
 
 1 
 
 13 
 
 8-8616 
 
 
 
 7 
 
 22-000 
 
 9 
 
 2 
 
 
 
 1 
 
 13 
 
 7-5341 
 
 
 
 7 
 
 21-375 
 
 9 
 
 1 
 
 7 
 
 1 
 
 13 
 
 6-2067 
 
 
 
 7 
 
 20-750 
 
 9 
 
 1 
 
 6 
 
 1 
 
 13 
 
 4-8793 
 
 
 
 7 
 
 20-125 
 
 9 
 
 1 
 
 5 
 
 1 
 
 13 
 
 3-5519 
 
 
 
 7 
 
 19-500 
 
 9 
 
 1 
 
 4 
 
 1 
 
 13 
 
 2-2245 
 
 
 
 7 
 
 19-875 
 
 9 
 
 1 
 
 3 
 
 1 
 
 13 
 
 0-8971 
 
 
 
 7 
 
 18-250 
 
 9 
 
 1 
 
 2 
 
 1 
 
 12 
 
 11-5697 
 
 
 
 7 
 
 17-625 
 
 9 
 
 1 
 
 1 
 
 1 
 
 12 
 
 10-2422 
 
 
 
 7 
 
 17-000 
 
 9 
 
 1 
 
 
 
 1 
 
 12 
 
 8-9168 
 
 
 
 7 
 
 16-375 
 
 9 
 
 
 
 7 
 
 1 
 
 12 
 
 7-5874 
 
 
 
 7 
 
 15-750 
 
 9 
 
 
 
 6 
 
 1 
 
 12 
 
 6-2600 
 
GOLD-VALUING TABLE. 
 
 Xlll 
 
 
 FINE 
 Per 
 
 GrOLD, 
 Ounce 
 
 CARAT G-OLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Ihuts. Grs. 
 
 Carats Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 7 
 
 15-125 
 
 9 
 
 
 
 5 
 
 1 
 
 12 
 
 4-9326 
 
 
 
 7 
 
 14-500 
 
 9 
 
 
 
 4 
 
 1 
 
 12 
 
 3-6052 
 
 
 
 7 
 
 13-875 
 
 9 
 
 
 
 3 
 
 1 
 
 12 
 
 2-2778 
 
 
 
 7 
 
 13-250 
 
 9 
 
 
 
 2 
 
 1 
 
 12 
 
 0-9503 
 
 
 
 7 
 
 12-625 
 
 9 
 
 
 
 1 
 
 1 
 
 11 
 
 11-6229 
 
 
 
 7 
 
 12-000 
 
 9 
 
 
 
 
 
 1 
 
 11 
 
 10-2954 
 
 
 
 7 
 
 11-375 
 
 8 
 
 3 
 
 7 
 
 1 
 
 11 
 
 8-9680 
 
 
 
 7 
 
 10-750 
 
 8 
 
 3 
 
 6 
 
 1 
 
 11 
 
 7-6406 
 
 
 
 7 
 
 10-125 
 
 8 
 
 3 
 
 5 
 
 1 
 
 11 
 
 6-3132 
 
 
 
 7 
 
 9-500 
 
 8 
 
 3 
 
 4 
 
 1 
 
 11 
 
 4-9857 
 
 
 
 7 
 
 8-875 
 
 8 
 
 3 
 
 3 
 
 1 
 
 11 
 
 3-6583 
 
 
 
 7 
 
 8-250 
 
 8 
 
 3 
 
 2 
 
 1 
 
 11 
 
 2-3309 
 
 
 
 7 
 
 7-625 
 
 8 
 
 3 
 
 1 
 
 1 
 
 11 
 
 1-0035 
 
 
 
 7 
 
 7-000 
 
 8 
 
 3 
 
 
 
 1 
 
 10 
 
 11-6761 
 
 
 
 7 
 
 6-375 
 
 8 
 
 2 
 
 7 
 
 1 
 
 10 
 
 10-3487 
 
 
 
 7 
 
 5-750 
 
 8 
 
 2 
 
 6 
 
 1 
 
 10 
 
 9-0213 
 
 
 
 7 
 
 5-125 
 
 8 
 
 2 
 
 5 
 
 1 
 
 10 
 
 7-6938 
 
 
 
 7 
 
 4-500 
 
 8 
 
 2 
 
 4 
 
 1 
 
 10 
 
 6-3664 
 
 
 
 7 
 
 ^3-875 
 
 8 
 
 2 
 
 3 
 
 1 
 
 10 
 
 5-0390 
 
 
 
 7 
 
 3-250 
 
 8 
 
 2 
 
 2 
 
 1 
 
 10 
 
 3-7116 
 
 
 
 7 
 
 2-625 
 
 8 
 
 2 
 
 1 
 
 1 
 
 10 
 
 2-3843 
 
 
 
 7 
 
 2-000 
 
 8 
 
 2 
 
 
 
 1 
 
 10 
 
 1-0568 
 
 
 
 7 
 
 1-375 
 
 8 
 
 1 
 
 7 
 
 1 
 
 9 
 
 11-7294 
 
 
 
 7 
 
 0-750 
 
 8 
 
 1 
 
 6 
 
 1 
 
 9 
 
 10-4019 
 
 
 
 7 
 
 0-125 
 
 8 
 
 1 
 
 5 
 
 1 
 
 9 
 
 9-0745 
 
 
 
 6 
 
 23-500 
 
 8 
 
 1 
 
 4 
 
 1 
 
 9 
 
 7-7471 
 
 
 
 6 
 
 22-875 
 
 8 
 
 1 
 
 3 
 
 1 
 
 9 
 
 6-4197 
 
 
 
 6 
 
 22-250 
 
 8 
 
 1 
 
 2 
 
 1 
 
 9 
 
 5-0923 
 
 
 
 6 
 
 21-625 
 
 8 
 
 1 
 
 1 
 
 1 
 
 9 
 
 3-7649 
 
 
 
 6 
 
 21-000 
 
 8 
 
 1 
 
 
 
 1 
 
 9 
 
 2-4375 
 
 
 
 6 
 
 20-375 
 
 8 
 
 
 
 7 
 
 1 
 
 9 
 
 1-1100 
 
 
 
 6 
 
 19-750 
 
 8 
 
 
 
 6 
 
 1 
 
 8 
 
 11-7826 
 
 
 
 6 
 
 19-125 
 
 8 
 
 
 
 5 
 
 1 
 
 8 
 
 10-4552 
 
 
 
 6 
 
 18-500 
 
 8 
 
 
 
 4 
 
 1 
 
 8 
 
 9-1278 
 
 
 
 6 
 
 17-875 
 
 8 
 
 
 
 3 
 
 1 
 
 8 
 
 7-8004 
 
 
 
 6 
 
 17-250 
 
 8 
 
 
 
 2 
 
 1 
 
 8 
 
 6-4730 
 
 
 
 6 
 
 16-625 
 
 8 
 
 
 
 1 
 
 1 
 
 8 
 
 5-1455 
 
 
 
 6 
 
 16-000 
 
 8 
 
 
 
 
 
 1 
 
 8 
 
 3-8181 
 
 
 
 6 
 
 15-375 
 
 7 
 
 3 
 
 7 
 
 1 
 
 8 
 
 2-4907 
 
 
 
 6 
 
 14-750 
 
 7 
 
 3 
 
 6 
 
 1 
 
 8 
 
 1-1633 
 
 
 
 6 
 
 14-125 
 
 7 
 
 3 
 
 5 
 
 1 
 
 7 
 
 11-8359 
 
 o 
 
 6 
 
 13-500 
 
 7 
 
 3 
 
 4 
 
 1 
 
 7 
 
 10-5085 
 
 
 
 6 
 
 12-875 
 
 7 
 
 3 
 
 3 
 
 1 
 
 7 
 
 9-1811 
 
 
 
 6 
 
 12-250 
 
 7 
 
 3 
 
 2 
 
 1 
 
 7 
 
 7-8536 
 
XIV 
 
 GOLD-VALUING TABLE. 
 
 
 FINE 
 Per 
 
 GOLD, 
 
 Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Dwts. GTS. 
 
 Carats 
 
 Grs. 
 
 Eighths 
 
 
 
 s. 
 
 A 
 
 
 
 6 
 
 11-625 
 
 7 
 
 3 
 
 1 
 
 I 
 
 7 
 
 6-5262 
 
 
 
 6 
 
 11-000 
 
 7 
 
 3 
 
 
 
 1 
 
 7 
 
 5-1988 
 
 
 
 6 
 
 10-375 
 
 7 
 
 2 
 
 7 
 
 1 
 
 7 
 
 3-8714 
 
 
 
 6 
 
 9-750 
 
 7 
 
 2 
 
 6 
 
 1 
 
 7 
 
 2-5440 
 
 
 
 6 
 
 9-125 
 
 7 
 
 2 
 
 5 
 
 1 
 
 7 
 
 1-2166 
 
 
 
 6 
 
 8-500 
 
 7 
 
 2 
 
 4 
 
 1 
 
 6 
 
 11-8892 
 
 
 
 6 
 
 7-875 
 
 7 
 
 2 
 
 3 
 
 1 
 
 6 
 
 10-5617 
 
 
 
 6 
 
 7-250 
 
 7 
 
 2 
 
 2 
 
 1 
 
 6 
 
 9-2343 
 
 
 
 6 
 
 6-625 
 
 7 
 
 2 
 
 1 
 
 1 
 
 6 
 
 7-8069 
 
 
 
 6 
 
 6-000 
 
 7 
 
 2 
 
 
 
 1 
 
 6 
 
 6-5795 
 
 
 
 6 
 
 5-375 
 
 7 
 
 1 
 
 7 
 
 1 
 
 6 
 
 5-2521 
 
 
 
 6 
 
 4-750 
 
 7 
 
 1 
 
 6 
 
 1 
 
 6 
 
 3-9247 
 
 
 
 6 
 
 4-125 
 
 7 
 
 1 
 
 5 
 
 1 
 
 6 
 
 2-5973 
 
 
 
 6 
 
 3-500 
 
 7 
 
 1 
 
 4 
 
 1 
 
 6 
 
 1-2698 
 
 
 
 6 
 
 2-875 
 
 7 
 
 1 
 
 3 
 
 1 
 
 5 
 
 11-9424 
 
 
 
 6 
 
 2-250 
 
 7 
 
 1 
 
 2 
 
 1 
 
 5 
 
 10-6150 
 
 
 
 6 
 
 1-625 
 
 7 
 
 1 
 
 1 
 
 1 
 
 5 
 
 9-2876 
 
 
 
 6 
 
 1-000 
 
 7 
 
 1 
 
 
 
 1 
 
 5 
 
 7-9602 
 
 
 
 6 
 
 0-375 
 
 7 
 
 
 
 7 
 
 1 
 
 5 
 
 6-6328 
 
 
 
 5 
 
 23-750 
 
 7 
 
 
 
 6 
 
 1 
 
 5 
 
 5-3054 
 
 
 
 5 
 
 23-125 
 
 7 
 
 
 
 5 
 
 1 
 
 5 
 
 3-9779 
 
 
 
 5 
 
 22-500 
 
 7 
 
 
 
 4 
 
 1 
 
 5 
 
 2-6505 
 
 
 
 5 
 
 21-875 
 
 7 
 
 
 
 3 
 
 1 
 
 5 
 
 1-3231 
 
 
 
 5 
 
 21-250 
 
 7 
 
 
 
 2 
 
 1 
 
 4 
 
 11-9957 
 
 
 
 5 
 
 20-625 
 
 7 
 
 
 
 1 
 
 1 
 
 4 
 
 10-6683 
 
 
 
 5 
 
 20-000 
 
 7 
 
 
 
 
 
 1 
 
 4 
 
 9-3409 
 
 
 
 5 
 
 19-375 
 
 6 
 
 3 
 
 7 
 
 1 
 
 4 
 
 8-0134 
 
 
 
 5 
 
 18-750 
 
 6 
 
 3 
 
 6 
 
 1 
 
 4 
 
 6-6860 
 
 
 
 5 
 
 18-125 
 
 6 
 
 3 
 
 5 
 
 1 
 
 4 
 
 5-3586 
 
 
 
 5 
 
 17-500 
 
 6 
 
 3 
 
 4 
 
 1 
 
 4 
 
 4-0312 
 
 
 
 5 
 
 16-875 
 
 6 
 
 3 
 
 3 
 
 1 
 
 4 
 
 2-7038 
 
 
 
 5 
 
 16-250 
 
 6 
 
 3 
 
 2 
 
 1 
 
 4 
 
 1-3764 
 
 
 
 5 
 
 15-625 
 
 6 
 
 3 
 
 1 
 
 1 
 
 4 
 
 0-0490 
 
 
 
 5 
 
 15-000 
 
 6 
 
 3 
 
 
 
 1 
 
 3 
 
 10-7216 
 
 
 
 5 
 
 14-375 
 
 6 
 
 2 
 
 7 
 
 1 
 
 3 
 
 9-3941 
 
 
 
 5 
 
 13-750 
 
 6 
 
 2 
 
 6 
 
 1 
 
 3 
 
 8-0667 
 
 
 
 5 
 
 13-125 
 
 6 
 
 2 
 
 5 
 
 1 
 
 3 
 
 6-7393 
 
 
 
 5 
 
 12-500 
 
 6 
 
 2 
 
 4 
 
 1 
 
 3 
 
 5-4119 
 
 
 
 5 
 
 11-875 
 
 6 
 
 2 
 
 3 
 
 i 
 
 3 
 
 4-0845 
 
 
 
 5 
 
 11-250 
 
 6 
 
 2 
 
 2 
 
 1 
 
 3 
 
 2-7571 
 
 
 
 5 
 
 10-625 
 
 6 
 
 2 
 
 1 > l 1 
 
 3 
 
 1-4297 
 
 
 
 5 
 
 10-000 
 
 6 
 
 2 
 
 
 
 1 
 
 3 
 
 0-1022 
 
 
 
 5 
 
 9-375 
 
 6 
 
 1 
 
 7 
 
 1 
 
 2 
 
 10-7748 
 
 i o 
 
 5 
 
 8-750 
 
 6 
 
 1 
 
 6 
 
 1 
 
 2 
 
 9-4474 
 
GOLD-VALUING TABLE. 
 
 XV 
 
 
 FINE 
 
 Per 
 
 GrOLD, 
 
 Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 to. 
 
 Dmts. Grs. 
 
 Carats Grs. 
 
 Eighths 
 
 s. 
 
 d. 
 
 
 
 5 
 
 8-125 
 
 6 
 
 1 
 
 5 
 
 I 2 
 
 8-1200 
 
 
 
 5 
 
 7-500 
 
 6 
 
 1 
 
 4 
 
 1 2 
 
 6-7926 
 
 
 
 5 
 
 6-875 
 
 6 
 
 1 
 
 3 
 
 1 2 
 
 5-4652 
 
 
 
 5 
 
 6-250 
 
 6 
 
 1 
 
 2 
 
 1 2 
 
 4-1377 
 
 
 
 5 
 
 5-625 
 
 6 
 
 1 
 
 1 
 
 1 2 
 
 2-8103 
 
 
 
 5 
 
 5-000 
 
 6 
 
 1 
 
 
 
 1 2 
 
 1-4829 
 
 
 
 5 
 
 4-375 
 
 6 
 
 
 
 7 
 
 1 2 
 
 0-1555 
 
 
 
 5 
 
 3-750 
 
 6 
 
 
 
 6 
 
 1 1 
 
 10-8281 
 
 
 
 5 
 
 3-125 
 
 6 
 
 
 
 5 
 
 1 1 
 
 9-5007 
 
 
 
 5 
 
 2-500 
 
 6 
 
 
 
 4 
 
 1 1 
 
 8-1733 
 
 
 
 5 
 
 1-875 
 
 6 
 
 
 
 3 
 
 1 1 
 
 6-8458 
 
 
 
 5 
 
 1-250 
 
 6 
 
 
 
 2 
 
 1 1 
 
 5-5184 
 
 
 
 5 
 
 0-625 
 
 6 
 
 
 
 1 
 
 1 1 
 
 4-1910 
 
 
 
 5 
 
 o-ooo 
 
 6 
 
 
 
 
 
 1 1 
 
 2-8636 
 
 
 
 4 
 
 23-375 
 
 5 
 
 3 
 
 7 
 
 1 1 
 
 1-5362 
 
 
 
 4 
 
 22-750 
 
 5 
 
 3 
 
 6 
 
 1 1 
 
 0-2088 
 
 
 
 4 
 
 22-125 
 
 5 
 
 3 
 
 5 
 
 1 
 
 10-8813 
 
 
 
 4 
 
 21-500 
 
 5 
 
 3 
 
 4 
 
 1 
 
 9-5539 
 
 
 
 4 
 
 20-875 
 
 5 
 
 3 
 
 3 
 
 1 
 
 8-2265 
 
 
 
 4 
 
 20*250 
 
 5 
 
 3 
 
 2 
 
 1 
 
 6-8991 
 
 
 
 4 
 
 19-625 
 
 5 
 
 3 
 
 1 
 
 1 
 
 5-5717 
 
 
 
 4 
 
 19-000 
 
 5 
 
 3 
 
 
 
 1 
 
 4-2443 
 
 
 
 4 
 
 18-375 
 
 5 
 
 2 
 
 7 
 
 1 
 
 2-9169 
 
 
 
 4 
 
 17-750 
 
 5 
 
 2 
 
 6 
 
 1 
 
 1-5894 
 
 
 
 4 
 
 17-125 
 
 5 
 
 2 
 
 5 
 
 1 
 
 0-2620 
 
 
 
 4 
 
 16-500 
 
 5 
 
 2 
 
 4 
 
 19 
 
 10-9346 
 
 
 
 4 
 
 15-875 
 
 5 
 
 2 
 
 3 
 
 19 
 
 9-6072 
 
 
 
 4 
 
 15-250 
 
 5 
 
 2 
 
 2 
 
 19 
 
 8-2798 
 
 
 
 4 
 
 14-625 
 
 5 
 
 2 
 
 1 
 
 19 
 
 6-9524 
 
 
 
 4 
 
 14-000 
 
 5 
 
 2 
 
 
 
 19 
 
 5-6250 
 
 
 
 4 
 
 13-375 
 
 5 
 
 1 
 
 7 
 
 19 
 
 4-2975 
 
 
 
 4 
 
 12-750 
 
 5 
 
 1 
 
 6 
 
 19 
 
 2-9701 
 
 
 
 4 
 
 12-125 
 
 5 
 
 1 
 
 5 
 
 19 
 
 1-6427 
 
 
 
 4 
 
 11-500 
 
 5 
 
 1 
 
 4 
 
 19 
 
 0-3153 
 
 
 
 4 
 
 10-875 
 
 5 
 
 1 
 
 3 
 
 18 
 
 10-9879 
 
 
 
 4 
 
 10-250 
 
 5 
 
 1 
 
 2 
 
 18 
 
 9-6605 
 
 
 
 4 
 
 9-625 
 
 5 
 
 1 
 
 1 
 
 18 
 
 8-3331 
 
 
 
 4 
 
 9-000 
 
 5 
 
 1 
 
 
 
 18 
 
 7-0056 
 
 
 
 4 
 
 8-375 
 
 5 
 
 
 
 7 
 
 18 
 
 . 5-6782 
 
 
 
 4 
 
 7-750 
 
 5 
 
 
 
 6 
 
 18 
 
 4-3508 
 
 
 
 - 4 
 
 7-125 
 
 5 
 
 
 
 5 
 
 18 
 
 3-0234 
 
 
 
 4 
 
 6-500 
 
 5 
 
 
 
 4 
 
 18 
 
 1-6960 
 
 
 
 4 
 
 5-875 
 
 5 
 
 
 
 3 
 
 18 
 
 0-3686 
 
 
 
 4 
 
 5-250 
 
 5 
 
 
 
 2 
 
 17 
 
 11-0411 
 
XVI 
 
 GOLD-VALUING TABLE. 
 
 FINE GrOLD, 
 
 Per Ounce 
 
 CARAT G-OLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Drvts 
 
 Grs. 
 
 Carats 
 
 Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 4 
 
 4-625 
 
 5 
 
 
 
 1 
 
 
 
 17 
 
 9-7137 
 
 
 
 4 
 
 4-000 
 
 5 
 
 
 
 
 
 
 
 17 
 
 8-3863 
 
 
 
 4 
 
 3-375 
 
 4 
 
 3 
 
 7 
 
 
 
 17 
 
 7-0589 
 
 
 
 4 
 
 2-750 
 
 4 
 
 3 
 
 6 
 
 
 
 17 
 
 5-7315 
 
 
 
 4 
 
 2-125 
 
 4 
 
 3 
 
 5 
 
 
 
 17 
 
 4-4041 
 
 
 
 4 
 
 1-500 
 
 4 
 
 3 
 
 4 
 
 
 
 17 
 
 3-0767 
 
 
 
 4 
 
 0-875 
 
 4 
 
 3 
 
 3 
 
 
 
 17 
 
 1-7492 
 
 
 
 4 
 
 0-250 
 
 4 
 
 3 
 
 2 
 
 
 
 17 
 
 0-4218 
 
 
 
 3 
 
 23-625 
 
 4 
 
 3 
 
 1 
 
 
 
 16 
 
 11-0944 
 
 
 
 3 
 
 23-000 
 
 4 
 
 3 
 
 
 
 
 
 16 
 
 9-7670 
 
 
 
 3 
 
 22-375 
 
 4 
 
 2 
 
 7 
 
 
 
 16 
 
 8-4396 
 
 
 
 3 
 
 21-750 
 
 4 
 
 2 
 
 6 
 
 
 
 16 
 
 7-1122 
 
 
 
 3 
 
 21-125 
 
 4 
 
 2 
 
 5 
 
 
 
 16 
 
 5-7848 
 
 
 
 3 
 
 20-500 
 
 4 
 
 2 
 
 4 
 
 
 
 16 
 
 4-4573 
 
 
 
 3 
 
 19-875 
 
 4 
 
 2 
 
 3 
 
 
 
 16 
 
 3-1299 
 
 
 
 3 
 
 19-250 
 
 4 
 
 2 
 
 2 
 
 
 
 16 
 
 1-8025 
 
 
 
 3 
 
 18-625 
 
 4 
 
 2 
 
 1 
 
 
 
 16 
 
 0-4751 
 
 
 
 3 
 
 18-000 
 
 4 
 
 2 
 
 
 
 
 
 15 
 
 11-1477 
 
 
 
 3 
 
 17-375 
 
 4 
 
 1 
 
 7 
 
 
 
 15 
 
 9-8203 
 
 
 
 3 
 
 16-750 
 
 4 
 
 1 
 
 6 
 
 
 
 15 
 
 8-4929 
 
 
 
 3 
 
 16-125 
 
 4 
 
 1 
 
 5 
 
 
 
 15 
 
 7-1655' 
 
 
 
 3 
 
 15-500 
 
 4 
 
 1 
 
 4 
 
 
 
 15 
 
 5-8380 
 
 
 
 3 
 
 14-875 
 
 4 
 
 1 
 
 3 
 
 
 
 15 
 
 4-5106 
 
 
 
 3 
 
 14-250 
 
 4 
 
 1 
 
 2 
 
 
 
 15 
 
 3-1832 
 
 
 
 3 
 
 13-625 
 
 4 
 
 1 
 
 1 
 
 
 
 15 
 
 1-8558 
 
 
 
 3 
 
 13-000 
 
 4 
 
 1 
 
 
 
 
 
 15 
 
 0-5284 
 
 
 
 3 
 
 12-375 
 
 4 
 
 
 
 7 
 
 
 
 14 
 
 11-2009 
 
 
 
 3 
 
 11-750 
 
 4 
 
 
 
 6 
 
 
 
 14 
 
 9-8735 
 
 
 
 3 
 
 11-125 
 
 4 
 
 
 
 5 
 
 
 
 14 
 
 8-5461 
 
 
 
 3 
 
 10-500 
 
 4 
 
 
 
 4 
 
 
 
 14 
 
 7-2187 
 
 
 
 3 
 
 9-875 
 
 4 
 
 
 
 3 
 
 
 
 14 
 
 5-8913 
 
 
 
 3 
 
 9-250 
 
 4 
 
 
 
 2 
 
 
 
 14 
 
 4-5639 
 
 
 
 3 
 
 8-625 
 
 4 
 
 
 
 1 
 
 
 
 14 
 
 3-2365 
 
 
 
 3 
 
 8-000 
 
 4 
 
 
 
 
 
 
 
 14 
 
 1-9090 
 
 
 
 3 
 
 7-375 
 
 3 
 
 3 
 
 7 
 
 
 
 14 
 
 0-5816 
 
 
 
 3 
 
 6-750 
 
 3 
 
 3 
 
 6 
 
 
 
 13 
 
 11-2542 
 
 
 
 3 
 
 6-125 
 
 3 
 
 3 
 
 5 
 
 
 
 13 
 
 9-9268 
 
 
 
 3 
 
 5-500 
 
 3 
 
 3 
 
 4 
 
 
 
 13 
 
 8-5994 
 
 
 
 3 
 
 4-875 
 
 3 
 
 3 
 
 3 
 
 
 
 13 
 
 7-2720 
 
 
 
 3 ' 
 
 4-250 
 
 3 
 
 3 
 
 2 
 
 
 
 13 
 
 5-9446 
 
 
 
 3 
 
 3-625 
 
 3 
 
 3 
 
 1 
 
 
 
 13 
 
 4-6171 
 
 
 
 3 
 
 3-000 
 
 3 
 
 3 
 
 
 
 
 
 13 
 
 3-2897 
 
 
 
 3 
 
 2-375 
 
 3 
 
 2 
 
 7 
 
 
 
 13 
 
 1-9623 
 
 
 
 3 
 
 1-750 
 
 3 
 
 2 
 
 6 
 
 
 
 13 
 
 0-6349 
 
GOLD -VALUING TABLE. 
 
 XV11 
 
 FINE GOLD, 
 
 Per Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Dwts. 
 
 r. ? . 
 
 Carats 
 
 r*. 
 
 Eighths 
 
 I 
 
 s. 
 
 d. 
 
 
 
 3 
 
 1-125 
 
 3 
 
 2 
 
 5 
 
 
 
 12 
 
 11-3075 
 
 
 
 3 
 
 0-500 
 
 3 
 
 2 
 
 4 
 
 
 
 12 
 
 9-9801 
 
 
 
 2 
 
 23-875 
 
 3 
 
 2 
 
 3 
 
 
 
 12 
 
 8-6527 
 
 
 
 2 
 
 23-250 
 
 3 
 
 2 
 
 2 
 
 
 
 12 
 
 7-3250 
 
 
 
 2 
 
 22-625 
 
 3 
 
 2 
 
 1 
 
 
 
 12 
 
 5-9978 
 
 
 
 2 
 
 22-000 
 
 3 
 
 2 
 
 
 
 
 
 12 
 
 4-6704 
 
 
 
 2 
 
 21-375 
 
 3 
 
 1 
 
 7 
 
 
 
 12 
 
 3-3430 
 
 
 
 2 
 
 20-750 
 
 3 
 
 1 
 
 6 
 
 
 
 12 
 
 2-0156 
 
 
 
 2 
 
 20-125 
 
 3 
 
 1 
 
 5 
 
 
 
 12 
 
 0-6882 
 
 
 
 2 
 
 19-500 
 
 3 
 
 1 
 
 4 
 
 
 
 11 
 
 11-3607 
 
 
 
 2 
 
 18-875 
 
 3 
 
 1 
 
 3 
 
 
 
 11 
 
 10-0333 
 
 
 
 2 
 
 18-250 
 
 3 
 
 1 
 
 2 
 
 
 
 11 
 
 8-7059 
 
 
 
 2 
 
 17-625 
 
 3 
 
 1 
 
 1 
 
 
 
 11 
 
 7-3785 
 
 
 
 2 
 
 17-000 
 
 3 
 
 1 
 
 
 
 
 
 11 
 
 6-0511 
 
 
 
 2 
 
 16-375 
 
 3 
 
 
 
 7 
 
 
 
 11 
 
 4-7237 
 
 
 
 2 
 
 15-750 
 
 3 
 
 
 
 6 
 
 
 
 11 
 
 3-3963 
 
 
 
 2 
 
 15-125 
 
 3 
 
 
 
 5 
 
 
 
 11 
 
 2-0688 
 
 
 
 2 
 
 14-500 
 
 3 
 
 
 
 4 
 
 
 
 11 
 
 0-7414 
 
 
 
 2 
 
 13-875 
 
 3 
 
 
 
 3 
 
 
 
 10 
 
 11-4140 
 
 
 
 2 
 
 13-250 
 
 3 
 
 
 
 2 
 
 
 
 10 
 
 10-0866 
 
 
 
 2 
 
 12-625 
 
 3 
 
 
 
 1 
 
 
 
 10 
 
 8-7592 
 
 
 
 2 
 
 12-000 
 
 3 
 
 
 
 
 
 
 
 10 
 
 7-4318 
 
 
 
 2 
 
 11-375 
 
 2 
 
 3 
 
 7 
 
 
 
 10 
 
 6-1044 
 
 
 
 2 
 
 10-750 
 
 2 
 
 3 
 
 6 
 
 
 
 10 
 
 4-7769 
 
 
 
 2 
 
 10-125 
 
 2 
 
 3 
 
 5 
 
 
 
 10 
 
 3-4495 
 
 
 
 2 
 
 9-500 
 
 2 
 
 3 
 
 4 
 
 
 
 10 
 
 2-1221 
 
 
 
 2 
 
 8-875 
 
 2 
 
 3 
 
 3 
 
 
 
 10 
 
 0-7947 
 
 
 
 2 
 
 8-250 
 
 2 
 
 3 
 
 2 
 
 
 
 9 
 
 11-4673 
 
 
 
 2 
 
 7-625 
 
 2 
 
 3 
 
 1 
 
 
 
 9 
 
 10-1399 
 
 
 
 2 
 
 7-000 
 
 2 
 
 3 
 
 
 
 
 
 9 
 
 8-8125 
 
 
 
 2 
 
 6-375 
 
 2 
 
 2 
 
 7 
 
 
 
 9 
 
 7-4850 
 
 
 
 2 
 
 5-750 
 
 2 
 
 2 
 
 6 
 
 
 
 9 
 
 6-1576 
 
 
 
 2 
 
 5-125 
 
 2 
 
 2 
 
 5 
 
 
 
 9 
 
 4-8302 
 
 
 
 2 
 
 4-500 
 
 2 
 
 2 
 
 4 
 
 
 
 9 
 
 3-5028 
 
 
 
 2 
 
 3-875 
 
 2 
 
 2 
 
 3 
 
 
 
 9 
 
 2-1754 
 
 
 
 2 
 
 3-250 
 
 2 
 
 2 
 
 2 
 
 
 
 9 
 
 0-8480 
 
 
 
 2 
 
 2-625 
 
 2 
 
 2 
 
 1 
 
 
 
 8 
 
 11-5205 
 
 
 
 2' 
 
 2-000 
 
 2 
 
 2 
 
 
 
 
 
 8 
 
 10-1931 
 
 
 
 2 
 
 1-375 
 
 2 
 
 1 
 
 7 
 
 
 
 8 
 
 8-8657 
 
 
 
 2 
 
 0-750 
 
 2 
 
 1 
 
 6 
 
 
 
 8 
 
 7-5383 
 
 
 
 2 
 
 0-125 
 
 2 
 
 1 
 
 5 
 
 
 
 8 
 
 6-2109 
 
 
 
 1 
 
 23-500 
 
 2 
 
 1 
 
 4 
 
 
 
 8 
 
 4-8835 
 
 
 
 1 
 
 22-875 
 
 2 
 
 1 
 
 3 
 
 
 
 8 
 
 3-5561 
 
 
 
 ] 
 
 22-250 
 
 2 
 
 1 
 
 2 
 
 
 
 8 
 
 2-2286 
 
 3 N 
 
XV111 
 
 GOLD -VALUING TABLE. 
 
 
 FINE 
 Per 
 
 GrOLD, 
 
 Ounce 
 
 CAKAT (TOLD, 
 
 Per Ounce 
 
 STERLING VALUE, 
 
 Per Ounce 
 
 Oz. 
 
 Dwts. Grs. 
 
 Carats 
 
 Grs. 
 
 Eighths 
 
 
 
 5. 
 
 d. 
 
 
 
 1 
 
 21-625 
 
 2 
 
 1 
 
 1 
 
 
 
 8 
 
 0-9012 
 
 
 
 1 
 
 21-000 
 
 2 
 
 1 
 
 
 
 
 
 7 
 
 11-5738 
 
 
 
 1 
 
 20-375 
 
 2 
 
 
 
 7 
 
 
 
 7 
 
 10-2464 
 
 
 
 1 
 
 19-750 
 
 2 
 
 
 
 6 
 
 
 
 7 
 
 8-9190 
 
 
 
 1 
 
 19-125 
 
 2 
 
 
 
 5 
 
 
 
 7 
 
 7-5916 
 
 
 
 1 
 
 18-500 
 
 2 
 
 
 
 4 
 
 
 
 7 
 
 6-2642 
 
 
 
 1 
 
 17-875 
 
 2 
 
 
 
 3 
 
 
 
 7 
 
 4-9367 
 
 
 
 1 
 
 17-250 
 
 2 
 
 
 
 2 
 
 
 
 7 
 
 3-6093 
 
 
 
 1 
 
 16-625 
 
 2 
 
 
 
 1 
 
 
 
 7 
 
 2-2819 
 
 
 
 1 
 
 16-000 
 
 2 
 
 
 
 
 
 
 
 7 
 
 0-9545 
 
 
 
 1 
 
 15-375 
 
 1 
 
 3 
 
 7 
 
 
 
 6 
 
 11-6271 
 
 
 
 1 
 
 14-750 
 
 1 
 
 3 
 
 6 
 
 
 
 6 
 
 10-2997 
 
 
 
 1 
 
 14-125 
 
 1 
 
 3 
 
 5 
 
 
 
 6 
 
 8-9723 
 
 
 
 1 
 
 13-500 
 
 1 
 
 3 
 
 4 
 
 
 
 6 
 
 7-6448 
 
 
 
 1 
 
 12-875 
 
 1 
 
 3 
 
 3 
 
 
 
 6 
 
 6-3174 
 
 
 
 1 
 
 12-250 
 
 1 
 
 3 
 
 2 
 
 
 
 6 
 
 4-9900 
 
 
 
 1 
 
 11-625 
 
 1 
 
 3 
 
 1 
 
 
 
 6 
 
 3-6626 
 
 
 
 1 
 
 11-000 
 
 1 
 
 3 
 
 
 
 
 
 6 
 
 2-3352 
 
 
 
 1 
 
 10-375 
 
 1 
 
 2 
 
 7 
 
 
 
 6 
 
 1-0078 
 
 
 
 1 
 
 9-75C 
 
 1 
 
 2 
 
 6 
 
 
 
 5 
 
 11-6803 
 
 
 
 1 
 
 9-125 
 
 1 
 
 2 
 
 5 
 
 
 
 5 
 
 10-3529 
 
 
 
 1 
 
 8-500 
 
 1 
 
 2 
 
 4 
 
 
 
 5 
 
 9-0255 
 
 
 
 1 
 
 7-875 
 
 1 
 
 2 
 
 3 
 
 
 
 5 
 
 7-6981 
 
 
 
 1 
 
 7-250 
 
 1 
 
 2 
 
 2 
 
 
 
 5 
 
 6-3707 
 
 
 
 1 
 
 6-625 
 
 1 
 
 2 
 
 1 
 
 
 
 5 
 
 5-0433 
 
 
 
 1 
 
 6-000 
 
 1 
 
 2 
 
 
 
 
 
 5 
 
 3-7159 
 
 
 
 ] 
 
 5-375 
 
 1 
 
 1 
 
 7 
 
 
 
 5 
 
 2-3884 
 
 
 
 1 
 
 4-750 
 
 1 
 
 1 
 
 6 
 
 
 
 5 
 
 1-0610 
 
 
 
 i 
 
 4-125 
 
 1 
 
 1 
 
 5 
 
 
 
 4 
 
 11-7336 
 
 
 
 1 
 
 3-500 
 
 1 
 
 1 
 
 4 
 
 
 
 4 
 
 10-4062 
 
 
 
 1 
 
 2-875 
 
 1 
 
 1 
 
 3 
 
 
 
 4 
 
 9-0788 
 
 
 
 1 
 
 2-250 
 
 1 
 
 1 
 
 2 
 
 
 
 4 
 
 7-7514 
 
 
 
 1 
 
 1-625 
 
 1 
 
 1 
 
 1 
 
 
 
 4 
 
 6-4240 
 
 
 
 1 
 
 1-000 
 
 1 
 
 1 
 
 
 
 
 
 4 
 
 5-0965 
 
 
 
 1 
 
 0-375 
 
 1 
 
 
 
 7 
 
 
 
 4 
 
 3-7691 
 
 
 
 
 
 23-750 
 
 1 
 
 
 
 6 
 
 
 
 4 
 
 2-4417 
 
 
 
 
 
 23-125 
 
 1 
 
 
 
 5 
 
 
 
 4 
 
 1-1143 
 
 
 
 
 
 22-500 
 
 1 
 
 
 
 4 
 
 
 
 3 
 
 11-7869 
 
 
 
 
 
 21-875 
 
 1 
 
 
 
 3 
 
 
 
 3 
 
 10-4595 
 
 
 
 
 
 21-250 
 
 1 
 
 
 
 2 
 
 
 
 3 
 
 9-1321 
 
 
 
 
 
 20-625 
 
 1 
 
 
 
 1 
 
 
 
 3 
 
 7-8046 
 
 
 
 
 
 20-000 
 
 1 
 
 
 
 
 
 
 
 3 
 
 6-4772 
 
 o 
 
 
 
 19-375 
 
 
 
 3 
 
 7 
 
 
 
 3 
 
 5-1498 
 
 
 
 
 
 18-750 
 
 
 
 3 
 
 6 
 
 
 
 3 
 
 3-8224 
 
(iOLD-VALUING TABLE. 
 
 XIX 
 
 
 
 FINE 
 
 Per 
 
 GrOLD, 
 
 Ounce 
 
 CARAT GOLD, 
 
 Per Ounce 
 
 STERLING VALUE 
 
 Per Ounce 
 
 Os. 
 
 Dwts. Grs. 
 
 Carats 
 
 Grs. 
 
 Eighths 
 
 
 
 s. 
 
 d. 
 
 
 
 C 
 
 18-125 
 
 
 
 3 
 
 5 
 
 
 
 3 
 
 2-4950 
 
 
 
 
 
 17-500 
 
 
 
 3 
 
 4 
 
 
 
 3 
 
 1-1676 
 
 
 
 
 
 16-875 
 
 
 
 3 
 
 3 
 
 
 
 2 
 
 11-8401 
 
 
 
 
 
 16-250 
 
 
 
 3 
 
 2 
 
 
 
 2 
 
 10-5127 
 
 
 
 
 
 15-625 
 
 
 
 3 
 
 1 
 
 
 
 2 
 
 9-1853 
 
 
 
 
 
 15-000 
 
 
 
 3 
 
 
 
 
 
 2 
 
 7-8579 
 
 
 
 
 
 14-375 
 
 
 
 2 
 
 7 
 
 
 
 2 
 
 6-5305 
 
 
 
 
 
 13-750 
 
 
 
 2 
 
 6 
 
 
 
 2 
 
 5-2031 
 
 
 
 
 
 13-125 
 
 
 
 2 
 
 5 
 
 
 
 2 
 
 3-8757 
 
 
 
 
 
 12-500 
 
 
 
 2 
 
 4 
 
 
 
 2 
 
 2-5482 
 
 
 
 
 
 11-875 
 
 
 
 2 
 
 3 
 
 
 
 2 
 
 1-2208 
 
 
 
 
 
 11-250 
 
 
 
 2 
 
 2 
 
 
 
 1 
 
 11-8934 
 
 
 
 
 
 10-625 
 
 
 
 2 
 
 1 
 
 
 
 1 
 
 10-5660 
 
 
 
 
 
 10-000 
 
 
 
 2 
 
 
 
 
 
 1 
 
 9-2386 
 
 
 
 
 
 9-375 
 
 
 
 1 
 
 7 
 
 
 
 1 
 
 7-9112 
 
 
 
 
 
 8-750 
 
 
 
 1 
 
 6 
 
 
 
 1 
 
 6-5838 
 
 
 
 
 
 8-125 
 
 
 
 1 
 
 5 
 
 
 
 1 
 
 5-2563 
 
 
 
 
 
 7-500 
 
 
 
 1 
 
 4 
 
 
 
 1 
 
 3-9289 
 
 
 
 
 
 6-875 
 
 
 
 1 
 
 3 
 
 
 
 1 
 
 2-6015 
 
 
 
 
 
 6-250 
 
 
 
 1 
 
 2 
 
 
 
 1 
 
 1-2741 
 
 
 
 
 
 5-625 
 
 
 
 1 
 
 L 
 
 
 
 
 
 11-9467 
 
 
 
 
 
 5-000 
 
 
 
 1 
 
 
 
 
 
 
 
 10-6193 
 
 
 
 
 
 4-375 
 
 
 
 
 
 7 
 
 
 
 
 
 9-2919 
 
 
 
 
 
 3-750 
 
 
 
 
 
 6 
 
 
 
 
 
 7-9644 
 
 
 
 
 
 3-125 
 
 
 
 
 
 5 
 
 
 
 
 
 6-6370 
 
 
 
 
 
 2-500 
 
 
 
 
 
 4 
 
 
 
 
 
 5-3096 
 
 
 
 
 
 1-875 
 
 
 
 
 
 3 
 
 
 
 
 
 3-9822 
 
 
 
 
 
 1-250 
 
 
 
 
 
 2 
 
 
 
 
 
 2-6548 
 
 
 
 
 
 0-625 
 
 
 
 
 
 1 
 
 
 
 
 
 1-3274 
 
XX 
 
 GOLD-VALUING TABLE. 
 
 To convert MINT VALUE into BANK VALUE when the Standard 
 is expressed in Carats, Grains, and Eighths. This can be 
 readily accomplished for every report by the following 
 Tables : 
 
 TABLE A. 
 
 CARATS 
 
 VALUE IN PENCE 
 
 CARATS 
 
 VALUE IK PENCE 
 
 1 
 
 0681 
 
 13 
 
 8863 
 
 2 
 
 1363 
 
 14 
 
 9545 
 
 3 
 
 2045 
 
 15 
 
 1-0227 
 
 4 
 
 2727 
 
 16 
 
 1-0909 
 
 5 
 
 3409 
 
 17 
 
 1-1590 
 
 6 
 
 4090 
 
 18 
 
 1-2272 
 
 7 
 
 4772 
 
 19 
 
 1-2954 
 
 8 
 
 5454 
 
 20 
 
 1-3636 
 
 9 
 
 6136 
 
 21 
 
 1-4318 
 
 10 
 
 6818 
 
 22 
 
 1-5000 
 
 11 
 
 7500 
 
 23 
 
 1-5681 
 
 12 
 
 8181 
 
 24 
 
 1-6363 
 
 TABLE B. 
 
 CAEAT GRAINS 
 
 VALUE IN PENCE 
 
 CARAT GRAINS 
 
 VALUE IN PESCE 
 
 1 
 2 
 
 0170 
 0340 
 
 3 
 
 4 
 
 0511 
 0681 
 
 TABLE C. 
 
 EIGHTH CARAT 
 GRAINS 
 
 VALUE IN PENCE 
 
 EIGHTH CARAT 
 GRAINS 
 
 VALUE IN PENCE 
 
 1 
 
 0021 
 
 5 
 
 0106 
 
 2 
 
 0042 
 
 6 
 
 0127 
 
 3 
 
 0063 
 
 7 
 
 0149 
 
 4 
 
 0085 
 
 8 
 
 0170 
 
GOLD-VALUING TABLE. 
 
 XXI 
 
 Table A gives the difference in price between Mint and Bank 
 value for each carat up to fine gold ; Table B the same for carat 
 grains ; and Table C the same for eighths of carat grains. 
 
 Now as the Bank value of gold is 3 17s. 9d. per oz. standard 
 against Mint value of 3 17s. lO^d., it follows by calculation 
 that fine gold would fetch, Bank price, only 4 4s. 9-8182d, 
 instead of 4 4s. ll-4545<i., as shown by Table I. of Mint 
 Values ; and the Bank value of 1 oz. of gold, of any standard 
 whatever, may be readily ascertained by the above Tables A, 
 B, and C, and Table I. the Tables A, B, and C, giving the 
 quantities in pence to be deducted from the corresponding stan- 
 dard in Table I. Thus, suppose it is necessary to ascertain the 
 Bank value of 1 oz. of gold of 14 carats 2 grains 5 eighths fine : 
 refer to Table A, at 14 carats is found *9545c. ; at 2 grains in 
 Table B is found -0340d. ; and at 5 eighths in Table C -0106d 
 Now -9545 + -0340 -f- -0106 = -9991, which has to be deducted 
 from 2 lls. 10-5575d. (see Table L), leaving 2 11s. 9'5564d. as 
 the Bank value of 1 oz. of gold of the above fineness. 
 
 TABLE II. 
 
 TABLE of relative proportions of FINE GOLD and ALLOY, with 
 the respective Mint Values of 1 oz. of each Alloy when the 
 Standard is expressed in Thousandths. 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 1000 
 
 ooo 
 
 4 4 11-4545 
 
 986 
 
 014 
 
 4 3 9-1821 
 
 999 
 
 001 
 
 4 4 10-4350 
 
 985 
 
 015 
 
 4 3 8-1627 
 
 998 
 
 002 
 
 4 4 9-4156 
 
 984 
 
 016 
 
 4 3 7-1432 
 
 997 
 
 003 
 
 4 4 8-3961 
 
 983 
 
 017 
 
 4 3 6-1238 
 
 996 
 
 004 
 
 4' 4 7-3767 
 
 982 
 
 018 
 
 4 3 5-1043 
 
 995 
 
 005 
 
 4 4 6-3572 
 
 981 
 
 019 
 
 4 3 4-0849 
 
 994 
 
 006 
 
 4 4 5-3378 
 
 980 
 
 020 
 
 4 3 3-0654 
 
 993 
 
 007 
 
 4 4 4-3183 
 
 979 
 
 021 
 
 4 3 2-0459 
 
 992 
 
 008 
 
 4 4 3-2989 
 
 978 
 
 022 
 
 4 3 1-0265 
 
 991 
 
 009 
 
 4 4 2-2793 
 
 977 
 
 023 
 
 4 3 0-0070 
 
 990 
 
 010 
 
 4 4 1-2600 
 
 976 
 
 024 
 
 4 2 10-9876 
 
 989 
 
 Oil 
 
 4 4 0-2405 
 
 975 
 
 025 
 
 4 2 9-9681 
 
 988 
 
 012 
 
 4 3 11-2210 
 
 974 
 
 026 
 
 4 2 8-9487 
 
 987 
 
 013 
 
 4 3 10-2016 
 
 973 
 
 027 
 
 4 2 7-9292 
 
XX11 
 
 GOLD-VALUING TABLE. 
 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 PINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 972 
 
 028 
 
 4 2 6-9098 
 
 929 
 
 071 
 
 3 18 11-0732 
 
 971 
 
 029 
 
 4 2 5-8903 
 
 928 
 
 072 
 
 3 18 10-0538 
 
 970 
 
 030 
 
 4 2 4-8709 
 
 927 
 
 073 
 
 3 18 9-0343 
 
 969 
 
 031 
 
 4 2 3-8504 
 
 9^6 
 
 074 
 
 3 18 8-0149 
 
 968 
 
 032 
 
 4 2 2-8319 
 
 925 
 
 075 
 
 3 18 6-9954 
 
 967 
 
 033 
 
 4 2 1-8125 
 
 924 
 
 076 
 
 3 18 5-9759 
 
 966 
 
 034 
 
 4 2 0-7930 
 
 923 
 
 077 
 
 3 18 4-9565 
 
 965 
 
 035 
 
 4 1 11-7736 
 
 922 
 
 078 
 
 3 18 3-9370 
 
 964 
 
 036 
 
 4 1 10-7541 
 
 921 
 
 079 
 
 3 18 2-9176 
 
 963 
 
 037 
 
 4 1 9-7347 
 
 920 
 
 080 
 
 3 18 1-8981 
 
 962 
 
 038 
 
 4 1 8-7152 
 
 919 
 
 081 
 
 3 18 0-8787 
 
 961 
 
 039 
 
 4 1 7-6958 
 
 918 
 
 082 
 
 3 17 11-8592 
 
 960 
 
 040 
 
 4 1 6-6763 
 
 917 
 
 083 
 
 3 17 10-8398 
 
 959 
 
 041 
 
 4 5-6569 
 
 916* 
 
 084 
 
 3 17 9-8203 
 
 958 
 
 042 
 
 4 1 4-6374 
 
 915 
 
 085 
 
 3 17 8-8009 
 
 957 
 
 043 
 
 4 3-6179 
 
 914 
 
 086 
 
 3 17 7-7814 
 
 956 
 
 044 
 
 4 2-5985 
 
 913 
 
 087 
 
 3 17 6-7619 
 
 955 
 
 045 
 
 4 1-5790 
 
 912 
 
 088 
 
 3 17 5-7425 
 
 954 
 
 046 
 
 4 1 0-5596 
 
 911 
 
 089 
 
 3 17 4-7230 
 
 953 
 
 047 
 
 4 11-5401 
 
 910 
 
 090 
 
 3 17 3-7036 
 
 952 
 
 048 
 
 4 10-5207 
 
 909 
 
 091 
 
 3 17 2-6841 
 
 951 
 
 049 
 
 4 9-5012 
 
 908 
 
 092 
 
 3 17 1-6647 
 
 950 
 
 050 
 
 4 8-4818 
 
 907 
 
 093 
 
 3 17 0-6452 
 
 949 
 
 051 
 
 4 7-4623 
 
 906 
 
 094 
 
 3 16 11-6258 
 
 948 
 
 052 
 
 4 6-4429 
 
 905 
 
 095 
 
 3 16 10-6063 
 
 947 
 
 053 
 
 4 5-4234 
 
 904 
 
 096 
 
 3 16 9-5869 
 
 946 
 
 054 
 
 4 4-4039 
 
 903 
 
 097 
 
 3 16 8-5674 
 
 945 
 
 055 
 
 4 3-3835 
 
 902 
 
 098 
 
 3 16 7-5479 
 
 944 
 
 056 
 
 4 2-3650 
 
 901 
 
 099 
 
 3 16 6-5285 
 
 943 
 
 057 
 
 4 1-3456 
 
 900 
 
 100 
 
 3 16 5-5090 
 
 942 
 
 058 
 
 4 0-3261 
 
 899 
 
 101 
 
 3 16 4-4896 
 
 941 
 
 059 
 
 3 19 11-3067 
 
 898 
 
 102 
 
 3 16 3-4701 
 
 940 
 
 060 
 
 3 19 10-2872 
 
 897 
 
 103 
 
 3 16 2-4507 
 
 939 
 
 061 
 
 3 19 9-2678 
 
 896 
 
 104 
 
 3 16 1-4312 
 
 938 
 
 062 
 
 3 19 8-2483 
 
 895 
 
 105 
 
 3 16 0-4118 
 
 937 
 
 063 
 
 3 19 7-2289 
 
 894 
 
 106 
 
 3 15 11-3923 
 
 936 
 
 064 
 
 3 19 6-2094 
 
 893 
 
 107 
 
 3 15 10-3729 
 
 935 
 
 065 
 
 3 19 5-1899 
 
 892 
 
 108 
 
 3 15 9-3534 
 
 934 
 
 066 
 
 3 19 4-1705 
 
 891 
 
 109 
 
 3 15 8-3339 
 
 933 
 
 067 
 
 3 19 3-1510 
 
 890 
 
 110 
 
 3 15 7-3145 
 
 932 
 
 068 
 
 3 19 2-1316 
 
 889 
 
 111 
 
 3 15 6-2950 
 
 931 
 
 069 
 
 3 19 1-1121 
 
 888 
 
 112 
 
 3 15 5-2756 
 
 930 
 
 070 
 
 3 19 0-0927 
 
 887 
 
 113 
 
 3 15 4-2561 
 
 * 916-666 Standard -083-333 3 17s. 10'5000<f. 
 
GOLD-VALUING TABLE. 
 
 XX111 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 8. d. 
 
 
 
 s. d. 
 
 886 
 
 114 
 
 3 15 3-2367 
 
 841 
 
 159 
 
 3 11 5-3612 
 
 885 
 
 115 
 
 3 15 2-2172 
 
 840 
 
 160 
 
 3 11 4-3418 
 
 884 
 
 116 
 
 3 15 1-1978 
 
 839 
 
 161 
 
 3 11 3-3223 
 
 883 
 
 117 
 
 3 15 0-1783 
 
 838 
 
 162 
 
 3 11 2-3029 
 
 882 
 
 118 
 
 3 14 11-1589 
 
 837 
 
 163 
 
 3 11 1-3834 
 
 881 
 
 119 
 
 3 14 10-1394 
 
 836 
 
 164 
 
 3 11 0-2639 
 
 880. 
 
 120 
 
 3 14 9-1199 
 
 835 
 
 165 
 
 3 10 11-2445 
 
 879 
 
 121 
 
 3 14 8-1005 
 
 834 
 
 166 
 
 3 10 10-2250 
 
 878 
 
 122 
 
 3 14 7-0810 
 
 833 
 
 167 
 
 3 10 9-2056 
 
 877 
 
 123 
 
 3 14 6-0616 
 
 832 
 
 168 
 
 3 10 8-1861 
 
 876 
 
 124 
 
 3 14 5-0421 
 
 831 
 
 169 
 
 3 10 7-1667 
 
 875 
 
 125 
 
 3 14 4-0227 
 
 830 
 
 170 
 
 3 10 6-1472 
 
 874 
 
 126 
 
 3 14 3-0032 
 
 829 
 
 171 
 
 3 10 5-1278 
 
 873 
 
 127 
 
 3 14 1-9838 
 
 828 
 
 172 
 
 3 10 4-1083 
 
 872 
 
 128 
 
 3 14 0-9643 
 
 827 
 
 173 
 
 3 10 3-0889 
 
 871 
 
 129 
 
 3 13 11-9449 
 
 826 
 
 174 
 
 3 10 2-0694 
 
 870 
 
 130 
 
 3 13 10-9254 
 
 825 
 
 175 
 
 3 10 1-0499 
 
 869 
 
 131 
 
 3 13 9-9059 
 
 824 
 
 176 
 
 3 10 0-0305 
 
 868 
 
 132 
 
 3 13 8-8865 
 
 823 
 
 177 
 
 3 9 11-0110 
 
 867 
 
 133 
 
 3 13 7-8670 
 
 822 
 
 178 
 
 3 9 9-9916 
 
 866 
 
 134 
 
 3 13 6-8476 
 
 821 
 
 179 
 
 3 9 8-9721 
 
 865 
 
 135 
 
 3 13 5-8281 
 
 820 
 
 180 
 
 3 9 7-9527 
 
 864 
 
 136 
 
 3 13 4-8087 
 
 819 
 
 181 
 
 3 9 6-9332 
 
 863 
 
 137 
 
 3 13 3-7892 
 
 818 
 
 182 
 
 3 9 5-9138 
 
 862 
 
 138 
 
 3 13 2-7698 
 
 817 
 
 183 
 
 3 9 4-8943 
 
 861 
 
 139 
 
 3 13 1-7503 
 
 816 
 
 184 
 
 3 9 3-8749 
 
 860 
 
 140 
 
 3 13 0-7309 
 
 815 
 
 185 
 
 3 9 2-8554 
 
 859 
 
 141 
 
 3 12 11-7114 
 
 814 
 
 186 
 
 3 9 1-8359 
 
 858 
 
 142 
 
 3 12 10-6919 
 
 813 
 
 187 
 
 3 9 0-8165 
 
 857 
 
 143 
 
 3 12 9-6725 
 
 812 
 
 188 
 
 3 8 11-7970 
 
 856 
 
 144 
 
 3 12 8-6530 
 
 811 
 
 189 
 
 3 8 10-7776 
 
 855 
 
 145 
 
 3 12 7-6336 
 
 810 
 
 190 
 
 3 8 9-7581 
 
 854 
 
 146 
 
 3 12 6-6141 
 
 809 
 
 191 
 
 3 8 8-7387 
 
 853 
 
 147 
 
 3 12 5-5947 
 
 808 
 
 192 
 
 3 8 7-7192 
 
 852 
 
 148 
 
 3 12'' 4-5752 
 
 807 
 
 193 
 
 3 8 6-6998 
 
 851 
 
 149 
 
 3 12 3-5558 
 
 806 
 
 194 
 
 3 8 5-6803 
 
 850 
 
 150 
 
 3 12 2-5363 
 
 805 
 
 195 
 
 3 8 4-6609 
 
 849 
 
 151 
 
 3 12 1-5169 
 
 804 
 
 196 
 
 3 8 3-6414 
 
 848 
 
 152 
 
 3 12 0-4974 
 
 803 
 
 197 
 
 3 8 2-6219 
 
 847 
 
 153 
 
 3 11 11-4779 
 
 802 
 
 198 
 
 3 8 1-6025 
 
 846 
 
 154 
 
 3 11 10-4585 
 
 801 
 
 199 
 
 3 8 0-5830 
 
 845 
 
 155 
 
 3 11 9-4390 
 
 800 
 
 200 
 
 3 7 11-5636 
 
 844 
 
 156 
 
 3 11 8-4196 
 
 799 
 
 201 
 
 3 7 10-5441 
 
 843 
 
 157 
 
 3 11 7-4001 
 
 798 
 
 202 
 
 3 7 9-5247 
 
 842 
 
 158 
 
 3 11 6-3807 
 
 797 
 
 203 
 
 3 7 8-5052 
 
XXIV 
 
 GOLD-VALUING TABLE. 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 I s. d. 
 
 
 
 s. d. 
 
 796 
 
 204 
 
 3 7 7-4858 
 
 751 
 
 249 
 
 3 3 9-6103 
 
 795 
 
 205 
 
 3 7 6-4663 
 
 750 
 
 250 
 
 3 3 8-5909 
 
 794 
 
 206 
 
 3 7 5-4469 
 
 749 
 
 251 
 
 3 3 7-5714 
 
 793 
 
 207 
 
 3 7 4-4274 
 
 748 
 
 252 
 
 3 3 6-5519 
 
 792 
 
 208 
 
 3 7 3-4979 
 
 747 
 
 253 
 
 3 3 5-5325 
 
 791 
 
 209 
 
 3 7 2-3885 
 
 746 
 
 254 
 
 3 3 4-5130 
 
 790 
 
 210 
 
 3 7 1-3690 
 
 745 
 
 255 
 
 3 3 3-4936 
 
 789 
 
 211 
 
 3 7 0-3496 
 
 744 
 
 256 
 
 3 3 2-4741 
 
 788 
 
 212 
 
 3 6 11-3301 
 
 743 
 
 257 
 
 3 3 1-4547 
 
 787 
 
 213 
 
 3 6 10-3107 
 
 742 
 
 258 
 
 3 3 0-4352 
 
 786 
 
 214 
 
 3 6 9-2912 
 
 741 
 
 -259 
 
 3 2 11-4158 
 
 785 
 
 215 
 
 3 6 8-2718 
 
 740 
 
 260 
 
 3 2 10-3963 
 
 784 
 
 216 
 
 3 6 7-2523 
 
 739 
 
 261 
 
 3 2 9-3769 
 
 783 
 
 217 
 
 3 6 6-2329 
 
 738 
 
 262 
 
 3 2 8-3574 
 
 782 
 
 218 
 
 3 6 5-2134 
 
 737 
 
 263 
 
 3 2 7-3379 
 
 781 
 
 219 
 
 3 6 4-1939 
 
 736 
 
 264 
 
 3 2 6-3185 
 
 780 
 
 220 
 
 3 6 3-1745 
 
 735 
 
 265 
 
 3 2 5-2990 
 
 779 
 
 221 
 
 3 6 2-1550 
 
 734 
 
 266 
 
 3 2 4-2796 
 
 778 
 
 222 
 
 3 6 1-1356 
 
 733 
 
 267 
 
 3 2 3-2601 
 
 777 
 
 223 
 
 3 6 0-1161 
 
 732 
 
 268 
 
 3 2 2-2407 
 
 776 
 
 224 
 
 3 5 11-0967 
 
 731 
 
 269 
 
 3 2 1-2212 
 
 775 
 
 225 
 
 3 5 10-0772 
 
 730 
 
 270 
 
 3 2 0-2018 
 
 774 
 
 226 
 
 3 5 9-0578 
 
 729 
 
 271 
 
 3 1 11-1823 
 
 773 
 
 227 
 
 3 5 8-0383 
 
 728 
 
 272 
 
 3 1 10-1629 
 
 772 
 
 228 
 
 3 5 7-0189 
 
 727 
 
 273 
 
 3 1 9-1434 
 
 771 
 
 229 
 
 3 5 5-9994 
 
 726 
 
 274 
 
 3 1 8-1239 
 
 770 
 
 230 
 
 3 5 4-9799 
 
 725 
 
 275 
 
 3 1 7-1045 
 
 769 
 
 231 
 
 3 5 3-9605 
 
 724 
 
 276 
 
 3 1 6-0850 
 
 768 
 
 232 
 
 3 5 2-9410 
 
 723 
 
 277 
 
 3 1 5-0656 
 
 767 
 
 233 
 
 3 5 1-9216 
 
 722 
 
 278 
 
 3 1 4-0461 
 
 766 
 
 234 
 
 3 5 0-9021 
 
 721 
 
 279 
 
 3 1 3-0267 
 
 765 
 
 235 
 
 3 4 11-8827 
 
 720 
 
 280 
 
 3 1 2-0072 
 
 764 
 
 236 
 
 3 4 10-8632 
 
 719 
 
 281 
 
 3 1 0-9878 
 
 763 
 
 237 
 
 3 4 9-8438 
 
 718 
 
 282 
 
 3 11-9683 
 
 762 
 
 238 
 
 3 4 8-8243 
 
 717 
 
 283 
 
 3 10-9489 
 
 761 
 
 239 
 
 3 4 7-8049 
 
 716 
 
 284 
 
 3 9-9294 
 
 760 
 
 240 
 
 3 4 6-7854 
 
 715 
 
 285 
 
 3 8-9099 
 
 759 
 
 241 
 
 3 4 5-7659 
 
 714 
 
 286 
 
 3 7-8905 
 
 758 
 
 242 
 
 3 4 4-7465 
 
 713 
 
 287 
 
 3 6-8710 
 
 757 
 
 243 
 
 3 4 3-7270 
 
 712 
 
 288 
 
 3 5-8516 
 
 756 
 
 244 
 
 3 4 2-7076 
 
 711 
 
 289 
 
 3 4-8321 
 
 755 
 
 245 
 
 3 4 1-6881 
 
 710 
 
 290 
 
 3 3-8127 
 
 754 
 
 246 
 
 3 4 0-6687 
 
 709 
 
 291 
 
 3 2-7932 
 
 753 
 
 247 
 
 3 3 11-6492 
 
 708 
 
 292 
 
 3 1-7738 
 
 752 
 
 248 
 
 3 3 10-6298 
 
 707 
 
 293 
 
 3 0-7543 
 
GOLD-VALUING TABLE. 
 
 XXV 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 706 
 
 294 
 
 2 19 11-7349 
 
 661 
 
 339 
 
 2 16 1-8594 
 
 705 
 
 295 
 
 2 19 10-7154 
 
 660 
 
 340 
 
 2 16 0-8399 
 
 704 
 
 296 
 
 2 19 9-6959 
 
 659 
 
 341 
 
 2 15 11-8205 
 
 703 
 
 297 
 
 2 19 8-6765 
 
 658 
 
 342 
 
 2 15 10-8010 
 
 702 
 
 298 
 
 2 19 7-6570 
 
 657 
 
 343 
 
 2 15 9-7816 
 
 701 
 
 299 
 
 2 19 6-6376 
 
 656 
 
 344 
 
 2 15 8-7621 
 
 700 
 
 300 
 
 2 ]9 5-6181 
 
 655 
 
 345 
 
 2 15 7-7427 
 
 699 
 
 301 
 
 2 19 4-5987 
 
 654 
 
 346 
 
 2 15 6-7232 
 
 698 
 
 302 
 
 2 19 3-5792 
 
 653 
 
 347 
 
 2 15 5-7038 
 
 697 
 
 303 
 
 2 19 2-5598 
 
 652 
 
 348 
 
 2 15 4-6843 
 
 696 
 
 304 
 
 2 19 1-5403 
 
 651 
 
 349 
 
 2 15 3-6649 
 
 695 
 
 305 
 
 2 19 0-5209 
 
 650 
 
 350 
 
 2 15 2-6454 
 
 694 
 
 306 
 
 2 18 11-5014 
 
 649 
 
 351 
 
 2 15 1-6259 
 
 693 
 
 307 
 
 2 18 10-4820 
 
 648 
 
 352 
 
 2 15 0-6065 
 
 692 
 
 308 
 
 2 18 9-4625 
 
 647 
 
 353 
 
 2 14 11-5870 
 
 691 
 
 309 
 
 2 18 8-4430 
 
 646 
 
 354 
 
 2 14 10-5676 
 
 690 
 
 310 
 
 2 18 7-4236 
 
 645 
 
 355 
 
 2 14 9-5481 
 
 689 
 
 311 
 
 2 18 6-4041 
 
 644 
 
 356 
 
 2 14 8-5287 
 
 688 
 
 312 
 
 2 18 5-3847 
 
 643 
 
 357 
 
 2 14 7-5092 
 
 687 
 
 313 
 
 2 18 4-3652 
 
 642 
 
 358 
 
 2 14 6-4898 
 
 686 
 
 314 
 
 2 18 3-3458 
 
 641 
 
 359 
 
 2 14 5-4703 
 
 685 
 
 315 
 
 2 18 2-3263 
 
 640 
 
 360 
 
 2 14 4-4509 
 
 684 
 
 316 
 
 2 18 1-3069 
 
 639 
 
 361 
 
 2 14 3-4314 
 
 683 
 
 317 
 
 2 18 0-2874 
 
 638 
 
 362 
 
 2 14 2-4120 
 
 682 
 
 318 
 
 2 17 11-2680 
 
 637 
 
 363 
 
 2 14 1-3925 
 
 681 
 
 319 
 
 2 17 10-2485 
 
 636 
 
 364 
 
 2 14 0-3730 
 
 680 
 
 320 
 
 2 17 9-2290 
 
 635 
 
 365 
 
 2 13 11-3536 
 
 679 
 
 321 
 
 2 17 8-2096 
 
 634 
 
 366 
 
 2 13 10-3341 
 
 678 
 
 322 
 
 2 17 7-1901 
 
 633 
 
 367 
 
 2 13 9-3147 
 
 677 
 
 323 
 
 2 17 6-1707 
 
 632 
 
 368 
 
 2 13 8-2952 
 
 676 
 
 324 
 
 2 17 5-1512 
 
 631 
 
 369 
 
 2 13 7-2758 
 
 675 
 
 325 
 
 2 17 4-1318 
 
 630 
 
 370 
 
 2 13 6-2563 
 
 674 
 
 326 
 
 2 17 3-1123 
 
 629 
 
 371 
 
 2 13 5-2369 
 
 673 
 
 327 
 
 2 17 2-0929 
 
 628 
 
 372 
 
 2 13 4-2174 
 
 672 
 
 328 
 
 2 17 1-0734 
 
 627 
 
 373 
 
 2 13 3-1979 
 
 671 
 
 329 
 
 2 17 0-0540 
 
 626 
 
 374 
 
 2 13 2-1785 
 
 670 
 
 330 
 
 2 16 11-0345 
 
 625 
 
 375 
 
 2 13 1-1590 
 
 669 
 
 331 
 
 2 16 10-0151 
 
 624 
 
 376 
 
 2 13 0-1396 
 
 668 
 
 332 
 
 2 16 8-9956 
 
 623 
 
 377 
 
 2 12 11-1201 
 
 667 
 
 333 
 
 2 16 7-9761 
 
 622 
 
 378 
 
 2 12 10-1007 
 
 666 
 
 334 
 
 2 16 6-9567 
 
 621 
 
 379 
 
 2 12 9-0812 
 
 665 
 
 335 
 
 2 16 5-9372 
 
 620 
 
 380 
 
 2 12 8-0618 
 
 664 
 
 336 
 
 2 16 4-9178 
 
 619 
 
 381 
 
 2 12 7-0423 
 
 663 
 
 337 
 
 2 16 3-8983 
 
 618 
 
 382 
 
 2 12 6-0229 
 
 662 
 
 338 
 
 2 16 2-8789 
 
 617 
 
 383 
 
 2 12 5-0034 
 
XXVI 
 
 GOLD- VALUING TABLE. 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 616 
 
 384 
 
 2 12 3-9839 
 
 571 
 
 429 
 
 2 8 6-1085 
 
 615 
 
 385 
 
 2 12 2-9645 
 
 570 
 
 430 
 
 2 8 5-0890 
 
 614 
 
 386 
 
 2 12 1-9451 
 
 569 
 
 431 
 
 2 8 4-0696 
 
 613 
 
 387 
 
 2 12 0-9256 
 
 568 
 
 432 
 
 2 8 3-0501 
 
 612 
 
 388 
 
 2 11 11-9061 
 
 567 
 
 433 
 
 2 8 2-0307 
 
 611 
 
 389 
 
 2 11 10-8867 
 
 566 
 
 434 
 
 2 8 1-0112 
 
 610 
 
 390 
 
 2 11 9-8672 
 
 565 
 
 435 
 
 2 7 11-9918 
 
 609 
 
 391 
 
 2 11 8-8478 
 
 564 
 
 436 
 
 2 7 10-9723 
 
 608 
 
 392 
 
 2 11 7-8283 
 
 563 
 
 437 
 
 2 7 9-9529 
 
 607 
 
 393 
 
 2 11 6-8089 
 
 562 
 
 438 
 
 2 7 8-9334 
 
 606 
 
 394 
 
 2 11 5-7894 
 
 561 
 
 439 
 
 2 7 7-9140 
 
 605 
 
 395 
 
 2 11 4-7699 
 
 560 
 
 440 
 
 2 7 6-8945 
 
 604 
 
 396 
 
 2 11 3-7505 
 
 559 
 
 441 
 
 2 7 5-8751 
 
 603 
 
 397 
 
 2 11 2-7311 
 
 558 
 
 442 
 
 2 7 4-8556 
 
 602 
 
 398 
 
 2 11 1-7116 
 
 557 
 
 443 
 
 2 7 3-8361 
 
 601 
 
 399 
 
 2 11 0-6921 
 
 556 
 
 444 
 
 2 7 2-8167 
 
 600 
 
 400 
 
 2 10 11-6727 
 
 555 
 
 445 
 
 2 7 1-7972 
 
 599 
 
 401 
 
 2 10 10-6532 
 
 554 
 
 446 
 
 2 7 0-7778 
 
 598 
 
 402 
 
 2 10 9-6338 
 
 553 
 
 447 
 
 2 6 11-7583 
 
 597 
 
 403 
 
 2 10 8-6143 
 
 552 
 
 448 
 
 2 6 10-7389 
 
 596 
 
 404 
 
 2 10 7-5949 
 
 551 
 
 449 
 
 2 6 9-7194 
 
 595 
 
 405 
 
 2 10 6-5754 
 
 550 
 
 450 
 
 2 6 8-6999 
 
 594 
 
 406 
 
 2 10 5-5559 
 
 549 
 
 451 
 
 2 6 7-6805 
 
 593 
 
 407 
 
 2 10 4-5365 
 
 548 
 
 452 
 
 2 6 6-6611 
 
 592 
 
 408 
 
 2 10 3-5170 
 
 547 
 
 453 
 
 2 6 5-6416 
 
 591 
 
 409 
 
 2 10 2-4976 
 
 546 
 
 454 
 
 2 6 4-6221 
 
 590 
 
 410 
 
 2 10 1-4781 
 
 545 
 
 455 
 
 2 6 3-6027 
 
 589 
 
 411 
 
 2 10 0-4587 
 
 544 
 
 456 
 
 2 6 2-5832 
 
 588 
 
 412 
 
 2 9 11-4392 
 
 543 
 
 457 
 
 2 6 1-5638 
 
 587 
 
 413 
 
 2 9 10-4198 
 
 542 
 
 458 
 
 2 6 0-5443 
 
 586 
 
 414 
 
 2 9 9-4003 
 
 541 
 
 459 
 
 2 5 11-5249 
 
 585 
 
 415 
 
 2 9 8-3809 
 
 540 
 
 460 
 
 2 5 10-5054 
 
 584 
 
 416 
 
 2 9 7-3614 
 
 539 
 
 461 
 
 2 5 9-4859 
 
 583 
 
 417 
 
 2 9 6-3419 
 
 538 
 
 462 
 
 2 5 8-4665 
 
 582 
 
 418 
 
 2 9 5-3225 
 
 537 
 
 463 
 
 2 5 7-4470 
 
 581 
 
 419 
 
 2 9 4-3030 
 
 536 
 
 464 
 
 2 5 6-4276 
 
 580 
 
 420 
 
 2 9 3-2836 
 
 535 
 
 465 
 
 2 5 5-4081 
 
 579 
 
 421 
 
 2 9 2-2641 
 
 534 
 
 466 
 
 2 5 4-3887 
 
 578 
 
 422 
 
 2 9 1-2447 
 
 533 
 
 467 
 
 2 5 3-3692 
 
 577 
 
 423 
 
 2 9 0-2252 
 
 532 
 
 468 
 
 2 5 2-3498 
 
 576 
 
 424 
 
 2 8 11-2058 
 
 531 
 
 469 
 
 2 5 1-3303 
 
 575 
 
 425 
 
 2 8 10-1863 
 
 530 
 
 470 
 
 2 5 0-3109 
 
 574 
 
 426 
 
 2 8 9-1669 
 
 529 
 
 471 
 
 2 4 11-2914 
 
 573 
 
 427 
 
 2 8 8-1474 
 
 528 
 
 472 
 
 2 4 10-2719 
 
 1 57? 
 
 428 
 
 2 8 7-1279 
 
 527 
 
 473 
 
 9 4 9-2525 
 
GOLD -VALUING TABLE. 
 
 XX VU 
 
 FINE 
 GOLD 
 
 ALLOT 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 526 
 
 474 
 
 2 4 8-2330 
 
 481 
 
 519 
 
 2 10-3576 
 
 525 
 
 475 
 
 2 4 7-2136 
 
 480 
 
 520 
 
 2 9-3381 
 
 524 
 
 476 
 
 2 4 6-1941 
 
 479 
 
 521 
 
 2 8-3187 
 
 523 
 
 477 
 
 2 4 5-1747 
 
 478 
 
 522 
 
 2 7-2992 
 
 522 
 
 478 
 
 2 4 4-1552 
 
 477 
 
 523 
 
 2 6-2798 
 
 521 
 
 479 
 
 2 4 3-1358 
 
 476 
 
 524 
 
 2 5-2603 
 
 520 
 
 480 
 
 2 4 2-1163 
 
 475 
 
 525 
 
 2 4-2409 
 
 519 
 
 481 
 
 2 4 1-0969 
 
 474 
 
 526 
 
 2 3-2214 
 
 518 
 
 482 
 
 2 4 0-0774 
 
 473 
 
 527 
 
 2 2-2020 
 
 517 
 
 483 
 
 2 3 11-0579 
 
 472 
 
 528 
 
 2 1-1825 
 
 516 
 
 484 
 
 2 3 10-0385 
 
 471 
 
 529 
 
 2 0-1630 
 
 515 
 
 485 
 
 2 3 9-0190 
 
 470 
 
 530 
 
 1 19 11-1436 
 
 514 
 
 486 
 
 2 3 7-9996 
 
 469 
 
 531 
 
 1 19 10-1241 
 
 513 
 
 487 
 
 2 3 6-9801 
 
 468 
 
 532 
 
 1 19 9-1047 
 
 512 
 
 488 
 
 2 3 5-9607 
 
 467 
 
 533 
 
 1 19 8-0852 
 
 511 
 
 489 
 
 2 3 4-9412 
 
 466 
 
 534 
 
 1 19 7-0658 
 
 510 
 
 490 
 
 2 3 3-9218 
 
 465 
 
 535 
 
 1 19 6-0463 
 
 509 
 
 491 
 
 2 3 2-9023 
 
 464 
 
 536 
 
 1 19 5-0269 
 
 508 
 
 492 
 
 2 3 1-8829 
 
 463 
 
 537 
 
 1 19 4-0074 
 
 507 
 
 493 
 
 2 3 0-8634 
 
 462 
 
 538 
 
 1 19 2-9879 
 
 506 
 
 494 
 
 2 2 11-8439 
 
 461 
 
 539 
 
 1 19 1-9685 
 
 505 
 
 495 
 
 2 2 10-8245 
 
 460 
 
 540 
 
 1 19 0-9490 
 
 504 
 
 496 
 
 2 2 9-8051 
 
 459 
 
 541 
 
 1 18 11-9296 
 
 503 
 
 497 
 
 2 2- 8-7856 
 
 458 
 
 542 
 
 1 18 10-9101 
 
 502 
 
 498 
 
 2 2 7-7661 
 
 457 
 
 543 
 
 1 18 9-8907 
 
 501 
 
 499 
 
 2 2 6-7467 
 
 456 
 
 544 
 
 1 18 8-8712 
 
 500 
 
 500 
 
 2 2 5-7272 
 
 455 
 
 545 
 
 1 18 7-8518 
 
 499 
 
 501 
 
 2 2 4-7078 
 
 454 
 
 546 
 
 1 18 6-8323 
 
 498 
 
 502 
 
 2 2 3-6883 
 
 453 
 
 547 
 
 1 18 5-8129 
 
 497 
 
 503 
 
 2 2 2-6689 
 
 452 
 
 548 
 
 1 18 4-7934 
 
 496 
 
 504 
 
 2 2 1-6494 
 
 451 
 
 549 
 
 1 18 3-7739 
 
 495 
 
 505 
 
 2 2 '' 0-6300 
 
 450 
 
 550 
 
 1 18 2-7545 
 
 494 
 
 506 
 
 2 1 11-6105 
 
 449 
 
 551 
 
 1 18 1-7351 
 
 493 
 
 507 
 
 2 1 10-5911 
 
 448 
 
 552 
 
 1 18 0-7156 
 
 492 
 
 508 
 
 2 1 9-5716 
 
 447 
 
 553 
 
 1 17 11-6961 
 
 491 
 
 509 
 
 2 1 8-5521 
 
 446 
 
 554 
 
 1 17 10-6767 
 
 490 
 
 510 
 
 2 1 7-5327 
 
 445 
 
 555 
 
 1 17 9-6572 
 
 489 
 
 511 
 
 2 1 1-5132 
 
 444 
 
 556 
 
 1 17 8-6378 
 
 488 
 
 512 
 
 2 1 5-4938 
 
 443 
 
 557 
 
 1 17 7-6183 
 
 487 
 
 513 
 
 2 1 4-4743 
 
 442 
 
 558 
 
 1 17 6-5989 
 
 486 
 
 514 
 
 2 1 3-4549 
 
 441 
 
 559 
 
 1 17 5-5794 
 
 485 
 
 515 
 
 2 1 2-4354 
 
 440 
 
 560 
 
 1 17 4-5599 
 
 484 
 
 516 
 
 2 1 1-4159 
 
 439 
 
 561 
 
 1 17 3-5405 
 
 483 
 
 517 
 
 2 1 0-3965 
 
 438 
 
 562 
 
 1 17 2-5211 
 
 482 
 
 518 
 
 2 11-3770 
 
 437 
 
 563 
 
 1 17 1-5016 
 
XXV111 
 
 GOLD-VALUING TABLE. 
 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 
 
 
 I 
 
 
 
 s. d. 
 
 
 
 t. d. 
 
 436 
 
 564 
 
 1 17 0-4821 
 
 391 
 
 609 
 
 1 13 2-6067 
 
 435 
 
 565 
 
 1 16 11-4627 
 
 390 
 
 610 
 
 1 13 1-5872 
 
 434 
 
 566 
 
 1 16 10-4432 
 
 389 
 
 611 
 
 1 13 0-5678 
 
 433 
 
 567 
 
 1 16 9-4238 
 
 388 
 
 612 
 
 1 12 11-5483 
 
 432 
 
 568 
 
 1 16 8-4043 
 
 387 
 
 613 
 
 1 12 10-5289 
 
 431 
 
 569 
 
 1 16 7-3849 
 
 386 
 
 614 
 
 1 12 9-5094 
 
 430 
 
 570 
 
 1 16 6-3654 
 
 385 
 
 615 
 
 1 12 8-4899 
 
 429 
 
 571 
 
 1 16 5-3459 
 
 384 
 
 616 
 
 1 12 7-4705 
 
 428 
 
 572 
 
 1 16 4-3265 
 
 383 
 
 617 
 
 1 12 6-4511 
 
 427 
 
 573 
 
 1 16 3-3070 
 
 382 
 
 618 
 
 1 12 5-4316 
 
 426 
 
 574 
 
 1 16 2-2876 
 
 381 
 
 619 
 
 1 12 4-4121 
 
 425 
 
 575 
 
 1 16 1-2681 
 
 380 
 
 620 
 
 1 12 3-3927 
 
 424 
 
 576 
 
 1 16 0-2487 
 
 379 
 
 621 
 
 1 12 2-3732 
 
 423 
 
 577 
 
 1 15 11-2292 
 
 378 
 
 622 
 
 1 12 1-3538 
 
 422 
 
 578 
 
 1 15 10-2098 
 
 377 
 
 623 
 
 1 12 0-3343 
 
 421 
 
 579 
 
 1 15 9-1903 
 
 376 
 
 624 
 
 1 11 11-3142 
 
 420 
 
 580 
 
 1 15 8-1709 
 
 375 
 
 625 
 
 1 11 10-2954 
 
 419 
 
 581 
 
 1 15 7-1514 
 
 374 
 
 626 
 
 1 11 9-2759 
 
 418 
 
 582 
 
 1 15 6-1319 
 
 373 
 
 627 
 
 1 11 8-2565 
 
 417 
 
 583 
 
 1 15 5-1125 
 
 372 
 
 628 
 
 1 11 7-2370 
 
 416 
 
 584 
 
 1 15 4-0930 
 
 371 
 
 629 
 
 1 11 6-2176 
 
 415 
 
 585 
 
 1 15 3-0736 
 
 370 
 
 630 
 
 1 11 5-1981 
 
 414 
 
 586 
 
 1 15 2-0541 
 
 369 
 
 631 
 
 1 11 4-1787 
 
 413 
 
 587 
 
 1 15 1-0347 
 
 368 
 
 632 
 
 1 11 3-1592 
 
 412 
 
 588 
 
 1 15 0-0152 
 
 367 
 
 -633 
 
 1 11 2-1398 
 
 411 
 
 589 
 
 1 14 10-9958 
 
 366 
 
 634 
 
 1 11 1-1203 
 
 410 
 
 590 
 
 1 14 9-9763 
 
 365 
 
 635 
 
 1 11 0-1009 
 
 409 
 
 591 
 
 1 14 8-9569 
 
 364 
 
 636 
 
 1 10 11-0814 
 
 408 
 
 592 
 
 1 14 7-9374 
 
 363 
 
 637 
 
 1 10 10-0620 
 
 407 
 
 593 
 
 1 14 6-9179 
 
 362 
 
 638 
 
 1 10 9-0425 
 
 406 
 
 594 
 
 1 14 5-8985 
 
 361 
 
 639 
 
 1 10 8-0230 
 
 405 
 
 595 
 
 1 14 4-8790 
 
 360 
 
 640 
 
 1 10 7-0036 
 
 404 
 
 596 
 
 1 14 3-8596 
 
 359 
 
 641 
 
 1 10 5-9841 
 
 403 
 
 597 
 
 1 14 2-8401 
 
 358 
 
 642 
 
 1 10 4-9647 
 
 402 
 
 598 
 
 1 14 1-8207 
 
 357 
 
 643 
 
 1 10 3-9452 
 
 401 
 
 599 
 
 1 14 0-8012 
 
 356 
 
 644 
 
 1 10 2-9258 
 
 400 
 
 .600 
 
 1 13 11-7818 
 
 355 
 
 645 
 
 1 10 1-9063 
 
 399 
 
 601 
 
 1 13 10-7623 
 
 354 
 
 646 
 
 1 10 0-8869 
 
 398 
 
 602 
 
 1 13 9-7429 
 
 353 
 
 647 
 
 1 9 11-8674 
 
 397 
 
 603 
 
 1 13 8-7234 
 
 352 
 
 648 
 
 1 9 10-8479 
 
 396 
 
 604 
 
 1 13 7-7039 
 
 351 
 
 649 
 
 1 9 9-8285 
 
 395 
 
 605 
 
 1 13 6-6845 
 
 350 
 
 650 
 
 1 9 8-8090 
 
 394 
 
 606 
 
 1 13 5-6651 
 
 349 
 
 651 
 
 1 9 7-7896 
 
 393 
 
 607 
 
 1 13 4-6456 
 
 348 
 
 652 
 
 1 9 6-7701 
 
 392 
 
 608 
 
 1 13 3-6261 
 
 347 
 
 653 
 
 1 9 5-7507 
 
GOLD- VALUING TABLE. 
 
 XXIX 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. a. 
 
 
 
 8. d. 
 
 346 
 
 654 
 
 1 9 4-7312 
 
 301 
 
 699 
 
 1 5 6-8558 
 
 345 
 
 655 
 
 1 9 3-7118 
 
 300 
 
 700 
 
 1 5 5-8363 
 
 344 
 
 656 
 
 1 9 2-6923 
 
 299 
 
 701 
 
 1 5 4-8169 
 
 343 
 
 657 
 
 1 9 1-6729 
 
 298 
 
 702 
 
 1 5 3-7974 
 
 342 
 
 658 
 
 1 9 0-6534 
 
 297 
 
 703 
 
 1 5 2-7779 
 
 341 
 
 659 
 
 1 8 11-6339 
 
 296 
 
 704 
 
 1 5 1-7585 
 
 340 
 
 660 
 
 1 8 10-6145 
 
 295 
 
 705 
 
 1 5 0-7390 
 
 339 
 
 661 
 
 1 8 9-5951 
 
 294 
 
 706 
 
 1 4 11-7196 
 
 338 
 
 662 
 
 1 8 8-5756 
 
 293 
 
 707 
 
 1 4 10-7011 
 
 337 
 
 663 
 
 1 8 7-5561 
 
 292 
 
 708 
 
 1 4 9-6807 
 
 336 
 
 664 
 
 1 8 6-5367 
 
 291 
 
 709 
 
 1 4 8-6612 
 
 335 
 
 665 
 
 1 8 5-5172 
 
 290 
 
 710 
 
 1 4 7-6418 
 
 334 
 
 666 
 
 1 8 4-4978 
 
 289 
 
 711 
 
 1 4 6-6223 
 
 333 
 
 667 
 
 1 8 3-4783 
 
 288 
 
 712 
 
 1 4 5-6029 
 
 332 
 
 668 
 
 1 8 2-4589 
 
 287 
 
 713 
 
 1 4 4-5834 
 
 331 
 
 669 
 
 1 8 1-4394 
 
 286 
 
 714 
 
 1 4 3-5639 
 
 330 
 
 670 
 
 1 8 0-4199 
 
 285 
 
 715 
 
 1 4 2-5445 
 
 329 
 
 671 
 
 1 7 11-4005 
 
 284 
 
 716 
 
 1 4 1-5251 
 
 328 
 
 672 
 
 1 7 10-3811 
 
 283 
 
 717 
 
 1 4 0-5056 
 
 327 
 
 673 
 
 1 7 9-3616 
 
 282 
 
 718 
 
 1 3 11-4861 
 
 326 
 
 674 
 
 1 7 8-3421 
 
 281 
 
 719 
 
 1 3 10-4667 
 
 325 
 
 675 
 
 1 7 7-3227 
 
 280 
 
 720 
 
 1 3 9-4472 
 
 324 
 
 676 
 
 1 7 6-3032 
 
 279 
 
 721 
 
 1 3 8-4278 
 
 323 
 
 677 
 
 1 7 5-2838 
 
 278 
 
 722 
 
 1 3 7-4083 
 
 322 
 
 678 
 
 1 7 4-2643 
 
 277 
 
 723 
 
 1 3 6-3889 
 
 321 
 
 679 
 
 1 7 3-2449 
 
 276 
 
 724 
 
 1 3 5-3694 
 
 320 
 
 680 
 
 1 7 2-2254 
 
 275 
 
 725 
 
 1 3 4-3499 
 
 319 
 
 681 
 
 1 7 1-2059 
 
 274 
 
 726 
 
 1 3 3-3305 
 
 318 
 
 682 
 
 1 7 0-1865 
 
 273 
 
 727 
 
 1 3 2-3110 
 
 317 
 
 683 
 
 1 6 11-1670 
 
 272 
 
 728 
 
 1 3 1-2916 
 
 316 
 
 684 
 
 1 6 1Q.-1476 
 
 271 
 
 729 
 
 1 3 0-2721 
 
 315 
 
 685 
 
 1 6 9-1281 
 
 270 
 
 730 
 
 1 2 11-2527 
 
 314 
 
 686 
 
 1 6 8-1087 
 
 269 
 
 731 
 
 1 2 10-2332 
 
 313 
 
 687 
 
 1 6 7-0892 
 
 268 
 
 732 
 
 1 2 9-2138 
 
 312 
 
 688 
 
 1 6 6-0698 
 
 267 
 
 733 
 
 1 2 8-1943 
 
 311 
 
 689 
 
 1 6 5-0503 
 
 266 
 
 234 
 
 1 2 7-1749 
 
 310 
 
 690 
 
 1 6 4-0309 
 
 265 
 
 735 
 
 1 2 6-1554 
 
 309 
 
 691 
 
 1 6 3-0114 
 
 264 
 
 736 
 
 1 2 5-1351 
 
 308 
 
 692 
 
 1 6 1-9919 
 
 263 
 
 737 
 
 1 2 4-1165 
 
 307 
 
 693 
 
 1 6 0-9725 
 
 262 
 
 738 
 
 1 2 3-0970 
 
 306 
 
 694 
 
 1 5 11-9530 
 
 261 
 
 739 
 
 1 2 2-0776 
 
 305 
 
 695 
 
 1 5 10-9336 
 
 260 
 
 740 
 
 1 2 1-0581 
 
 304 
 
 696 
 
 1 5 9-9141 
 
 259 
 
 741 
 
 1 2 0-0387 
 
 303 
 
 697 
 
 1 5 8-8947 
 
 258 
 
 742 
 
 1 1 11-0192 
 
 302 
 
 698 
 
 1 5 7-8752 
 
 257 
 
 743 
 
 1 1 9-9998 
 
XXX 
 
 GOLD-VALUING TABLE. 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 256 
 
 744 
 
 1 8-9803 
 
 211 
 
 789 
 
 17 11-1049 
 
 255 
 
 745 
 
 1 7-9609 
 
 210 
 
 790 
 
 17 10-0854 
 
 254 
 
 746 
 
 1 6-9414 
 
 209 
 
 791 
 
 17 9-0659 
 
 253 
 
 747 
 
 1 5-9219 
 
 208 
 
 792 
 
 17 8-0465 
 
 252 
 
 748 
 
 1 4-9025 
 
 207 
 
 793 
 
 17 7-0270 
 
 251 
 
 749 
 
 1 3-8830 
 
 206 
 
 794 
 
 17 6-0076 
 
 250 
 
 750 
 
 1 2-8636 
 
 205 
 
 795 
 
 17 4-9881 
 
 249 
 
 751 
 
 1 1 1-8441 
 
 204 
 
 796 
 
 17 3-9687 
 
 248 
 
 752 
 
 1 1 0-8247 
 
 203 
 
 797 
 
 17 2-9492 
 
 247 
 
 753 
 
 1 11-8052 
 
 202 
 
 798 
 
 17 1-9298 
 
 246 
 
 754 
 
 1 10-7858 
 
 201 
 
 799 
 
 17 0-9103 
 
 245 
 
 755 
 
 1 9-7663 
 
 200 
 
 800 
 
 16 11-8909 
 
 244 
 
 756 
 
 1 8-7469 
 
 199 
 
 801 
 
 16 10-8714 
 
 243 
 
 757 
 
 1 7-7274 
 
 198 
 
 802 
 
 16 9-8519 
 
 242 
 
 758 
 
 1 6-7079 
 
 197 
 
 803 
 
 16 8-8325 
 
 241 
 
 759 
 
 1 5-6885 
 
 196 
 
 804 
 
 16 7-8130 
 
 240 
 
 760 
 
 1 4-6690 
 
 195 
 
 805 
 
 16 6-7936 
 
 239 
 
 761 
 
 1 3-6496 
 
 194 
 
 806 
 
 16 5-7741 
 
 238 
 
 762 
 
 1 2-6301 
 
 193 
 
 807 
 
 16 4-7547 
 
 237 
 
 763 
 
 1 1-6107 
 
 192 
 
 808 
 
 16 3-7352 
 
 236 
 
 764 
 
 1 0-5912 
 
 191 
 
 809 
 
 16 2-7158 
 
 235 
 
 765 
 
 19 11-5718 
 
 190 
 
 810 
 
 16 1-6963 
 
 234 
 
 766 
 
 19 10-5523 
 
 189 
 
 811 
 
 16 0-6769 
 
 233 
 
 767 
 
 19 9-5329 
 
 188 
 
 812 
 
 15 11-6574 
 
 232 
 
 768 
 
 19 8-5134 
 
 187 
 
 813 
 
 15 10-6379 
 
 231 
 
 769 
 
 19 7-4939 
 
 186 
 
 814 
 
 15 9-6185 
 
 230 
 
 770 
 
 19 6-4745 
 
 185 
 
 815 
 
 15 8-5990 
 
 229 
 
 771 
 
 19 5-4551 
 
 184 
 
 816 
 
 15 7-5796 
 
 228 
 
 772 
 
 19 4-4356 
 
 183 
 
 817 
 
 15 6-5601 
 
 227 
 
 773 
 
 19 3-4161 
 
 182 
 
 818 
 
 15 5-5407 
 
 226 
 
 774 
 
 19 2-3967 
 
 181 
 
 819 
 
 15 4-5212 
 
 225 
 
 775 
 
 19 1-3772 
 
 180 
 
 820 
 
 15 3-5018 
 
 224 
 
 776 
 
 19 0-3578 
 
 179 
 
 821 
 
 15 2-4823 
 
 223 
 
 777 
 
 18 11-3383 
 
 178 
 
 822 
 
 15 1-4629 
 
 222 
 
 778 
 
 18 10-3189 
 
 177 
 
 823 
 
 15 0-4434 
 
 221 
 
 779 
 
 18 9-2994 
 
 176 
 
 824 
 
 14 11-4239 
 
 220 
 
 780 
 
 18 8-2799 
 
 175 
 
 825 
 
 14 10-4045 
 
 219 
 
 781 
 
 18 7-2605 
 
 174 
 
 826 
 
 14 9-3851 
 
 218 
 
 782 
 
 18 6-2410 
 
 173 
 
 827 
 
 14 8-3656 
 
 217 
 
 783 
 
 18 5-2216 
 
 172 
 
 828 
 
 14 7-3461 
 
 216 
 
 784 
 
 18 4-2021 
 
 171 
 
 829 
 
 14 6-3267 
 
 215 
 
 785 
 
 18 3-L827 
 
 170 
 
 830 
 
 14 5-3072 
 
 214 
 
 786 
 
 18 2-1632 
 
 169 
 
 831 
 
 14 4-2878 
 
 213 
 
 787 
 
 18 1-1438 
 
 168 
 
 832 
 
 14 3-2683 
 
 212 
 
 788 
 
 18 0-1243 
 
 167 
 
 833 
 
 14 2-2489 
 
GOLD-VALUING TABLE. 
 
 XXXI 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 166 
 
 834 
 
 14 1-2294 
 
 121 
 
 879 
 
 10 3-3530 
 
 165 
 
 835 
 
 14 0-2099 
 
 120 
 
 880 
 
 10 2-3345 
 
 164 
 
 836 
 
 13 11-1905 
 
 119 
 
 881 
 
 10 1-3151 
 
 163 
 
 837 
 
 13 10-1710 
 
 118 
 
 882 
 
 10 0-2956 
 
 162 
 
 838 
 
 13 9-1516 
 
 117 
 
 883 
 
 9 11-2761 
 
 161 
 
 839 
 
 13 8-1321 
 
 116 
 
 884 
 
 9 10-2567 
 
 160 
 
 840 
 
 13 7-1127 
 
 115 
 
 885 
 
 9 9-2372 
 
 159 
 
 841 
 
 13 6-0932 
 
 114 
 
 886 
 
 9 8-2178 
 
 158 
 
 842 
 
 13 5-0738 
 
 113 
 
 887 
 
 9 7-1983 
 
 157 
 
 843 
 
 13 4-0543 
 
 112 
 
 888 
 
 9 6-1789 
 
 156 
 
 844 
 
 13 3-0349 
 
 111 
 
 889 
 
 9 5-1594 
 
 155 
 
 845 
 
 13 2-0154 
 
 110 
 
 890 
 
 9 4-1399 
 
 154 
 
 846 
 
 13 0-9959 
 
 109 
 
 891 
 
 9 3-1205 
 
 153 
 
 847 
 
 12 11-9765 
 
 108 
 
 892 
 
 9 2-1010 
 
 152 
 
 848 
 
 12 10-9570 
 
 107 
 
 893 
 
 9 1-0816 
 
 151 
 
 849 
 
 12 9-9376 
 
 106 
 
 894 
 
 9 0-0621 
 
 150 
 
 850 
 
 12 8-9181 
 
 105 
 
 895 
 
 8 11-0427 
 
 149 
 
 851 
 
 12 7-8987 
 
 104 
 
 896 
 
 8 10-0232 
 
 148 
 
 852 
 
 12 6-8792 
 
 103 
 
 897 
 
 8 9-0038 
 
 147 
 
 853 
 
 12 5-8598 
 
 102 
 
 898 
 
 8 7-9843 
 
 146 
 
 854 
 
 12 4-8403 
 
 101 
 
 899 
 
 8 6-9649 
 
 145 
 
 855 
 
 12 3-8209 
 
 100 
 
 900 
 
 8 5-9454 
 
 144 
 
 856 
 
 12 2-8014 
 
 99 
 
 901 
 
 8 4-9259 
 
 143 
 
 857 
 
 12 1-7819 
 
 98 
 
 902 
 
 8 3-9065 
 
 142 
 
 858 
 
 12 0-7625 
 
 97 
 
 903 
 
 8 2-8870 
 
 141 
 
 859 
 
 11 11-7430 
 
 96 
 
 904 
 
 8 1-8676 
 
 140 
 
 860 
 
 11 10-7236 
 
 95 
 
 905 
 
 8 0-8481 
 
 139 
 
 861 
 
 11 9-7041 
 
 94 
 
 906 
 
 7 11-8287 
 
 138 
 
 862 
 
 11 8-6847 
 
 93 
 
 907 
 
 7 10-8092 
 
 137 
 
 863 
 
 11 7-6652 
 
 92 
 
 908 
 
 7 9-7898 
 
 136 
 
 864 
 
 11 6-6458 
 
 91 
 
 909 
 
 7 8-7703 
 
 135 
 
 865 
 
 11 5-6263 
 
 90 
 
 910 
 
 ? 7-7509 
 
 134 
 
 866 
 
 11 4-6069 
 
 89 
 
 911 
 
 7 6-7314 
 
 133 
 
 867 
 
 11 3-5874 
 
 88 
 
 912 
 
 7 5-7119 
 
 132 
 
 868 
 
 11 2-5679 
 
 87 
 
 913 
 
 7 4-6925 
 
 131 
 
 869 
 
 11 1-5485 
 
 86 
 
 914 
 
 7 3-6730 
 
 130 
 
 870 
 
 11 0-5290 
 
 85 
 
 915 
 
 7 2-6536 
 
 129 
 
 871 
 
 10 11-5096 
 
 84 
 
 916 
 
 7 1-6341 
 
 128 
 
 872 
 
 10 10-4901 
 
 83 
 
 917 
 
 7 0-6147 
 
 127 
 
 873 
 
 10 9-4707 
 
 82 
 
 918 
 
 6 11-5952 
 
 126 
 
 874 
 
 10 8-4512 
 
 81 
 
 919 
 
 6 10-5758 
 
 125 
 
 875 
 
 10 7-4318 
 
 80 
 
 920 
 
 6 9-5563 
 
 124 
 
 876 
 
 10 6-4123 
 
 79 
 
 921 
 
 6 8-5369 
 
 123 
 
 877 
 
 10 5-3929 
 
 78 
 
 722 
 
 6 7-5174 
 
 122 
 
 878 
 
 10 4-3734 
 
 77 
 
 923 
 
 6 6-4979 
 
XXX11 
 
 GOLD-VALUING TABLE. 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 FINE 
 GOLD 
 
 ALLOY 
 
 VALUE 
 
 
 
 s. d. 
 
 
 
 s. d. 
 
 76 
 
 924 
 
 6 5-4785 
 
 38 
 
 962 
 
 3 2-7392 
 
 75 
 
 925 
 
 6 4-4590 
 
 37 
 
 963 
 
 3 1-7198 
 
 74 
 
 926 
 
 6 3-4396 
 
 . 36 
 
 964 
 
 3 0-7003 
 
 73 
 
 927 
 
 6 2-4201 
 
 35 
 
 965 
 
 2 11-6809 
 
 72 
 
 928 
 
 6 1-4007 
 
 34 
 
 966 
 
 2 10-6614 
 
 71 
 
 929 
 
 6 0-3812 
 
 33 
 
 967 
 
 2 9-6419 
 
 70 
 
 930 
 
 5 11-3618 
 
 32 
 
 968 
 
 2 8-6225 
 
 69 
 
 931 
 
 5 10-3423 
 
 31 
 
 969 
 
 2 7-6030 
 
 68 
 
 932 
 
 5 9-3229 
 
 30 
 
 970 
 
 2 6-5836 
 
 67 
 
 933 
 
 5 8-3034 
 
 29 
 
 971 
 
 2 5-5641 
 
 66 
 
 934 
 
 5 7-2839 
 
 28 
 
 972 
 
 2 4-5447 
 
 65 
 
 935 
 
 5 6-2645 
 
 27 
 
 973 
 
 2 3-5252 
 
 64 
 
 936 
 
 5 5-2451 
 
 26 
 
 974 
 
 2 2-5058 
 
 63 
 
 937 
 
 5 4-2256 
 
 25 
 
 975 
 
 2 1-4863 
 
 62 
 
 938 
 
 5 3-2061 
 
 24 
 
 976 
 
 2 0-4669 
 
 61 
 
 939 
 
 5 2-1867 
 
 23 
 
 977 
 
 1 11-4474 
 
 60 
 
 940 
 
 5 1-1672 
 
 22 
 
 978 
 
 1 10-4279 
 
 59 
 
 941 
 
 5 0-1478 
 
 21 
 
 979 
 
 1 9-4085 
 
 58 
 
 942 
 
 4 11-1283 
 
 20 
 
 980 
 
 1 8-3890 
 
 57 
 
 943 
 
 4 10-1089 
 
 19 
 
 981 
 
 1 7-3696 
 
 56 
 
 944 
 
 4 9-0894 
 
 18 
 
 982 
 
 1 6-3501 
 
 55 
 
 945 
 
 4 8-0699 
 
 17 
 
 983 
 
 1 5-3307 
 
 54 
 
 946 
 
 4 7-0505 
 
 16 
 
 984 
 
 1 4-3112 
 
 53 
 
 947 
 
 4 6-0310 
 
 15 
 
 985 
 
 1 3-2918 
 
 52 
 
 948 
 
 4 5-0116 
 
 14 
 
 986 
 
 1 2-2723 
 
 51 
 
 949 
 
 4 3-9921 
 
 13 
 
 987 
 
 1 1-2529 
 
 50 
 
 950 
 
 4 2-9727 
 
 12 
 
 988 
 
 1 0-2334 
 
 49 
 
 951 
 
 4 1-9532 
 
 11 
 
 989 
 
 11-2139 
 
 48 
 
 952 
 
 4 0-9338 
 
 10 
 
 990 
 
 10-1945 
 
 47 
 
 953 
 
 3 11-9143 
 
 9 
 
 991 
 
 9-1750 
 
 46 
 
 954 
 
 3 10-8949 
 
 8 
 
 992 
 
 8-1556 
 
 45 
 
 955 
 
 3 9-8754 
 
 7 
 
 993 
 
 7-1361 
 
 44 
 
 956 
 
 3 8-8559 
 
 6 
 
 994 
 
 6-1167 
 
 43 
 
 957 
 
 3 7-8365 
 
 5 
 
 995 
 
 5-0972 
 
 42 
 
 958 
 
 3 6-8170 
 
 4 
 
 996 
 
 4-0778 
 
 41 
 
 959 
 
 3 5-7976 
 
 3 
 
 997 
 
 3-0583 
 
 40 
 
 960 
 
 3 4-7781 
 
 2 
 
 998 
 
 2-0389 
 
 39 
 
 961 
 
 3 3-7587 
 
 1 
 
 999 
 
 1-0194 
 
GOLD-VALUING TABLE. 
 
 XXX111 
 
 To convert MINT VALUE into BANK VALUE when the Standard is 
 expressed in Thousandths. 
 
 Thousandths 
 
 Value in Pence 
 
 Thousandths 
 
 Value in Pence 
 
 1 
 
 001636 
 
 6 
 
 009816 . 
 
 2 
 
 003272 
 
 7 
 
 011352 
 
 3 
 
 004908 
 
 8 
 
 013088 
 
 4 
 
 006544 
 
 9 
 
 014724 
 
 5 
 
 008180 
 
 
 
 To illustrate the use of the above table, gold of ^^ths fine 
 may be taken. As in the Table for finding the Bank value of 
 gold when the standard is reported in carats, &c., the amounts 
 in pence, as above, are to be deducted from the prices attached 
 to corresponding standards in Table No. 2. Thus, the minus 
 value of ^ 5 O ths is '00818 of a penny ; therefore, the minus value 
 of T 5_o_o_ths is *818 of a penny, which amount must be deducted 
 from the Mint price of gold at the above standard. On refer- 
 ring to the Table it will be found to be 2 2s. 5-7272d. per oz. 
 Now, if -818 be deducted, the' remainder will be 2 2s. 4'9092d., 
 representing the Bank value of 1 oz. of gold of the fineness just 
 mentioned. 
 
 3o 
 
XXXI V 
 
 ASSAY TABLE. 
 
 TABLE III. 
 
 ASSAY TABLE, showing the Amount of GOLD or SILVER, in Ounces, 
 Pennyweights, and Grains, contained in a Ton of Ore, &c. 
 from the Weight of Metal obtained in an Assay of 200 Grains 
 of Mineral. 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 Ore give of 
 
 will yield of 
 
 Ore give of 
 
 will yield of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 001 
 
 036 
 
 031 
 
 516 
 
 002 
 
 6 12 
 
 032 
 
 5 4 12 
 
 003 
 
 9 19 
 
 033 
 
 5 7 19 
 
 004 
 
 13 1 
 
 034 
 
 5 11 1 
 
 005 
 
 16 8 
 
 035 
 
 5 14 8 
 
 006 
 
 19 14 
 
 036 
 
 5 17 14 
 
 007 
 
 1 2 20 
 
 037 
 
 6 20 
 
 008 
 
 1 6 3 
 
 038 
 
 643 
 
 009 
 
 1 9 9 
 
 -039 
 
 679 
 
 010 
 
 1 12 6 
 
 040 
 
 6 10 16 
 
 Oil 
 
 1 15 22 
 
 041 
 
 6 13 22 
 
 012 
 
 1 19 4 
 
 042 
 
 6 17 4 
 
 013 
 
 2211 
 
 043 
 
 7 11 
 
 014 
 
 2 5 17 
 
 044 
 
 7 3 17 
 
 015 
 
 290 
 
 045 
 
 770 
 
 016 
 
 2 12 6 
 
 046 
 
 7 10 6 
 
 017 
 
 2 15 12 
 
 047 
 
 7 13 12 
 
 018 
 
 2 18 19 
 
 048 
 
 7 10 19 
 
 019 
 
 321 
 
 049 
 
 8 1 
 
 020 
 
 358 
 
 050 
 
 838 
 
 021 
 
 3 8 14 
 
 051 
 
 8 6 14 
 
 022 
 
 3 11 20 
 
 052 
 
 8 9 20 
 
 023 
 
 3 15 3 
 
 053 
 
 8 13 3 
 
 024 
 
 3 18 9 
 
 054 
 
 8 16 9 
 
 025 
 
 4 1 16 
 
 055 
 
 8 19 16 
 
 026 
 
 4 4 22 
 
 056 
 
 9 2 22 
 
 027 
 
 484 
 
 057 
 
 964 
 
 028 
 
 4 11 11 
 
 058 
 
 9 9 11 
 
 029 
 
 4 14 17 
 
 059 
 
 9 12 17 
 
 030 
 
 4 18 
 
 060 
 
 9 16 
 
ASSAY TABLE. 
 
 XXXV 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 Ore give of 
 
 will yield of 
 
 Ore give of 
 
 will yield of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Or. 
 
 Oz. Dwts. Grs. 
 
 6V. 
 
 Oz. Dwts. Grs. 
 
 061 
 
 9 19 6 
 
 105 
 
 17 3 
 
 062 
 
 10 2 12 
 
 106 
 
 17 6 6 
 
 063 
 
 10 5 19 
 
 107 
 
 17 9 12 
 
 064 
 
 10 9 1 
 
 108 
 
 17 12 19 
 
 065 
 
 10 12 8 
 
 109 
 
 17 16 1 
 
 066 
 
 10 15 14 
 
 110 
 
 17 19 8 
 
 067 
 
 10 18 20 
 
 111 
 
 18 2 14 
 
 068 
 
 11 2 3 
 
 112 
 
 18 5 20 
 
 069 
 
 11 5 9 
 
 113 
 
 18 9 3 
 
 070 
 
 11 8 16 
 
 114 
 
 18 12 9 
 
 071 
 
 11 11 22 
 
 115 
 
 18 15 16 
 
 072 
 
 11 15 4 
 
 116 
 
 18 18 22 
 
 073 
 
 11 18 11 
 
 117 
 
 19 2 4 
 
 074 
 
 12 1 17 
 
 118 
 
 19 5 11 
 
 075 
 
 12 5 
 
 119 
 
 19 8 17 
 
 076 
 
 12 8 6 
 
 120 
 
 19 12 
 
 077 
 
 12 11 12 
 
 121 
 
 19 15 6 
 
 078 
 
 12 14 19 
 
 122 
 
 19 18 12 
 
 079 
 
 12 18 1 
 
 123 
 
 20 1 19 
 
 080 
 
 13 1 8 
 
 124 
 
 20 5 1 
 
 081 
 
 13 4 14 
 
 125 
 
 20 8 8 
 
 082 
 
 13 7 20 
 
 126 
 
 20 11 14 
 
 083 
 
 13 11 3 
 
 127 
 
 20 14 20 
 
 084 
 
 13 14 9 
 
 128 
 
 20 18 3 
 
 085 
 
 13 17 16 
 
 129 
 
 21 1 9 
 
 086 
 
 14 22 
 
 130 
 
 21 4 16 
 
 087 
 
 14 4 4 
 
 131 
 
 21 7 22 
 
 088 
 
 14 7 11 
 
 132 
 
 21 11 4 
 
 089 
 
 14 10 17 
 
 133 
 
 21 14 11 
 
 090 
 
 14 14 
 
 134 
 
 21 17 17 
 
 091 
 
 14 17 6 
 
 135 
 
 22 1 
 
 092 
 
 15 12 
 
 136 
 
 22 4 6 
 
 093 
 
 15 3 19 
 
 137 
 
 22 7 12 
 
 094 
 
 15 7 1 
 
 138 
 
 22 10 19 
 
 095 
 
 15 10 8 
 
 139 
 
 22 14 1 
 
 096 
 
 15 13 14 
 
 140 
 
 22 17 8 
 
 097 
 
 15 16 20 
 
 141 
 
 23 14 
 
 098 
 
 16 3 
 
 142 
 
 23 3 20 
 
 099 
 
 16 3 9 
 
 143 
 
 23 7 3 
 
 100 
 
 16 6 16 
 
 144 
 
 23 10 9 
 
 101 
 
 16 9 22 
 
 145 
 
 23 13 16 
 
 "102 
 
 16 13 4 
 
 146 
 
 23 16 22 
 
 103 
 
 16 16 11 
 
 147 
 
 24 4 
 
 104 
 
 15 19 17 
 
 148 
 
 24 3 11 
 
 3 o 2 
 
XXXVI 
 
 ASSAY TABLE. 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 Ore give of 
 
 will yield of 
 
 Ore give of 
 
 will yield of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 149 
 
 24 6 17 
 
 193 
 
 31 10 11 
 
 150 
 
 24 10 
 
 194 
 
 31 13 17 
 
 151 
 
 24 13 6 
 
 195 
 
 31 17 
 
 152 
 
 24 16 12 
 
 196 
 
 32 6 
 
 153 
 
 24 19 19 
 
 197 
 
 32 3 12 
 
 154 
 
 25 3 1 
 
 198 
 
 32 6 19 
 
 155 
 
 25 6 8 
 
 199 
 
 32 10 1 
 
 156 
 
 25 9 14 
 
 200 
 
 32 13 8 
 
 157 
 
 25 12 20 
 
 201 
 
 32 16 14 
 
 158 
 
 25 16 3 
 
 202 
 
 32 19 20 
 
 159 
 
 25 19 9 
 
 203 
 
 33 3 3 
 
 160 
 
 26 2 16 
 
 204 
 
 33 6 9 
 
 161 
 
 26 5 22 
 
 205 
 
 33 9 16 
 
 162 
 
 26 9 4 
 
 206 
 
 33 12 22 
 
 163 
 
 26 12 11 
 
 207 
 
 33 16 4 
 
 164 
 
 26 15 17 
 
 208 
 
 33 19 11 
 
 165 
 
 26 19 
 
 209 
 
 34 2 17 
 
 166 
 
 27 2 6 
 
 210 
 
 34 6 
 
 167 
 
 27 5 12 
 
 211 
 
 34 9 6 
 
 168 
 
 27 8 19 
 
 212 
 
 34 12 12 
 
 169 
 
 27 12 1 
 
 213 
 
 34 15 19 
 
 170 
 
 27 15 8 
 
 214 
 
 34 19 1 
 
 171 
 
 27 18 14 
 
 215 
 
 35 2 8 
 
 172 
 
 28 1 20 
 
 216 
 
 35 5 14 
 
 173 
 
 28 5 3 
 
 217 
 
 35 8 20 
 
 174 
 
 28 8 9 
 
 218 
 
 35 12 3 
 
 175 
 
 28 11 16 
 
 219 
 
 35 15 9 
 
 176 
 
 28 14 22 
 
 220 
 
 35 18 16 
 
 177 
 
 28 18 4 
 
 221 
 
 36 1 22 
 
 178 
 
 29 1 11 
 
 222 
 
 36 5 4 
 
 179 
 
 29 4 17 
 
 223 
 
 36 8 11 
 
 180 
 
 29 8 
 
 224 
 
 36 11 17 
 
 181 
 
 29 11 6 
 
 225 
 
 36 15 
 
 182 
 
 29 14 12 
 
 226 
 
 36 18 6 
 
 183 
 
 29 17 19 
 
 227 
 
 37 1 12 
 
 184 
 
 30 1 1 
 
 228 
 
 37 4 19 
 
 185 
 
 30 4 8 
 
 229 
 
 37 8 1 
 
 186 
 
 30 7 14 
 
 230 
 
 37 11 8 
 
 187 
 
 30 10 20 
 
 231 
 
 37 14 14 
 
 188 
 
 30 14 3 
 
 232 
 
 37 17 20 
 
 189 
 
 30 17 9 
 
 233 
 
 38 1 3 
 
 190 
 
 31 16 
 
 234 
 
 38 4 9 
 
 191 
 
 31 3 22 
 
 235 
 
 38 7 16 
 
 192 
 
 31 7 4 
 
 236 
 
 38 10 22 
 
ASSAY TABLE. 
 
 XXXV11 
 
 If 200 Grains of 
 
 One 
 
 Ton of Ore 
 
 If 200 Grains of 
 
 One 
 
 Ton of Ore 
 
 Ore give of 
 
 wi 
 
 11 yield 
 
 of 
 
 Ore give of 
 
 will yield 
 
 of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 237 
 
 38 
 
 14 
 
 4 
 
 281 
 
 45 
 
 17 
 
 22 
 
 238 
 
 38 
 
 17 
 
 11 
 
 282 
 
 46 
 
 1 
 
 4 
 
 239 
 
 39 
 
 
 
 17 
 
 283 
 
 46 
 
 4 
 
 11 
 
 240 
 
 39 
 
 4 
 
 
 
 284 
 
 46 
 
 7 
 
 17 
 
 241 
 
 39 
 
 7 
 
 6 
 
 285 
 
 46 
 
 11 
 
 
 
 242 
 
 39 
 
 10 
 
 12 
 
 286 
 
 46 
 
 14 
 
 6 
 
 243 
 
 39 
 
 13 
 
 18 
 
 287 
 
 46 
 
 17 
 
 12 
 
 244 
 
 39 
 
 17 
 
 1 
 
 288 
 
 47 
 
 
 
 19 
 
 245 
 
 40 
 
 
 
 8 
 
 289 
 
 47 
 
 4 
 
 1 
 
 246 
 
 40 
 
 3 
 
 14 
 
 290 
 
 47 
 
 7 
 
 8 
 
 247 
 
 40 
 
 6 
 
 20 
 
 291 
 
 47 
 
 10 
 
 14 
 
 248 
 
 40 
 
 10 
 
 3 
 
 292 
 
 47 
 
 13 
 
 20 
 
 249 
 
 40 
 
 13 
 
 9 
 
 293 
 
 47 
 
 17 
 
 3 
 
 250 
 
 40 
 
 16 
 
 16 
 
 294 
 
 48 
 
 
 
 9 
 
 251 
 
 40 
 
 19 
 
 22 
 
 295 
 
 48 
 
 3 
 
 16 
 
 252 
 
 41 
 
 3 
 
 4 
 
 296 
 
 48 
 
 6 
 
 22 
 
 253 
 
 41 
 
 6 
 
 11 
 
 297 
 
 48 
 
 10 
 
 4 
 
 254 
 
 41 
 
 9 
 
 17 
 
 298 
 
 48 
 
 13 
 
 11 
 
 255 
 
 41 
 
 13 
 
 
 
 299 
 
 48 
 
 16 
 
 17 
 
 256 
 
 41 
 
 16 
 
 6 
 
 300 
 
 49 
 
 
 
 
 
 257 
 
 41 
 
 19 
 
 12 
 
 301 
 
 49 
 
 3 
 
 6 
 
 258 
 
 42 
 
 2 
 
 19 
 
 302 
 
 49 
 
 6 
 
 12 
 
 259 
 
 42 
 
 6 
 
 1 
 
 303 
 
 49 
 
 9 
 
 19 
 
 260 
 
 42 
 
 9 
 
 8 
 
 304 
 
 49 
 
 13 
 
 1 
 
 261 
 
 42 
 
 12 
 
 14 
 
 305 
 
 49 
 
 16 
 
 8 
 
 262 
 
 42 
 
 15 
 
 20 
 
 306 
 
 49 
 
 19 
 
 14 
 
 263 
 
 42 
 
 19 
 
 3 
 
 307 
 
 50 
 
 2 
 
 20 
 
 264 
 
 43 
 
 2 
 
 9 
 
 308 
 
 50 
 
 6 
 
 3 
 
 265 
 
 43 
 
 5 
 
 16 
 
 309 
 
 50 
 
 9 
 
 9 
 
 266 
 
 43 
 
 8 
 
 22 
 
 310 
 
 50 
 
 12 
 
 16 
 
 267 
 
 43 
 
 12 
 
 4 
 
 311 
 
 50 
 
 15 
 
 22 
 
 268 
 
 43 
 
 15 
 
 11 
 
 312 
 
 50 
 
 19 
 
 4 
 
 269 
 
 43 
 
 18 
 
 17 
 
 313 
 
 51 
 
 2 
 
 11 
 
 270 
 
 44 
 
 2 
 
 
 
 314 
 
 51 
 
 5 
 
 17 
 
 271 
 
 44 
 
 5 
 
 6 
 
 315 
 
 51 
 
 9 
 
 
 
 272 
 
 44 
 
 8 
 
 12 
 
 316 
 
 51 
 
 12 
 
 6 
 
 273 
 
 44 
 
 11 
 
 19 
 
 317 
 
 51 
 
 15 
 
 12 
 
 274 
 
 44 
 
 15 
 
 1 
 
 318 
 
 51 
 
 18 
 
 19 
 
 275 
 
 44 
 
 18 
 
 8 
 
 319 
 
 52 
 
 2 
 
 1 
 
 276 
 
 45 
 
 1 
 
 14 
 
 320 
 
 52 
 
 5 
 
 8 
 
 277 
 
 45 
 
 4 
 
 20 
 
 321 
 
 52 
 
 8 
 
 14 
 
 278 
 
 45 
 
 8 
 
 3 
 
 322 
 
 52 
 
 11 
 
 20 
 
 279 
 
 45 
 
 11 
 
 9 
 
 323 
 
 52 
 
 15 
 
 3 
 
 280 
 
 45 
 
 14 
 
 16 
 
 324 
 
 52 
 
 18 
 
 9 
 
XXXV111 
 
 ASSAY TABLE. 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 If 200 Grains 
 
 of One Ton of Ore 
 
 Ore give of 
 
 will yield of 
 
 Ore give of 
 
 will yield of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE MKTAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 325 
 
 53 1 16 
 
 369 
 
 60 5 9 
 
 326 
 
 53 4 22 
 
 370 
 
 60 8 16 
 
 327 
 
 53 8 4 
 
 371 
 
 60 11 22 
 
 328 
 
 53 11 11 
 
 372 
 
 60 15 4 
 
 329 
 
 53 14 17 
 
 373 
 
 60 18 11 
 
 330 
 
 53 18 
 
 374 
 
 61 1 17 
 
 331 
 
 54 1 6 
 
 375 
 
 61 5 
 
 332 
 
 54 4 12 
 
 376 
 
 61 8 6 
 
 333 
 
 54 7 19 
 
 377 
 
 61 11 12 
 
 334 
 
 54 11 1 
 
 378 
 
 61 14 19 
 
 335 
 
 54 14 8 
 
 379 
 
 61 18 1 
 
 336 
 
 54 17 14 
 
 380 
 
 62 1 8 
 
 337 
 
 55 20 
 
 381 
 
 62 4 14 
 
 338 
 
 55 4 3 
 
 382 
 
 62 7 20 
 
 339 
 
 55 7 9 
 
 383 
 
 62 11 3 
 
 340 
 
 55 10 16 
 
 384 
 
 62 14 9 
 
 341 
 
 55 13 22 
 
 385 
 
 62 17 16 
 
 342 
 
 55 17 4 
 
 386 
 
 63 22 
 
 343 
 
 56 11 
 
 387 
 
 63 4 4 
 
 344 
 
 56 317 
 
 388 
 
 63 7 11 
 
 345 
 
 56 7 
 
 389 
 
 63 10 17 
 
 346 
 
 56 10 6 
 
 390 
 
 63 14 
 
 347 
 
 56 13 12 
 
 391 
 
 63 17 6 
 
 348 
 
 56 16 19 
 
 392 
 
 64 12 
 
 349 
 
 57 1 
 
 393 
 
 64 3 19 
 
 350 
 
 57 38 
 
 394 
 
 64 7 1 
 
 351 
 
 57 6 14 
 
 395 
 
 64 10 8 
 
 352 
 
 57 9 20 
 
 396 
 
 64 13 14 
 
 353 
 
 57 13 3 
 
 397 
 
 . 64 16 20 
 
 354 
 
 57 16 9 
 
 398 
 
 65 3 
 
 355 
 
 57 19 16 
 
 399 
 
 65 3 9 
 
 356 
 
 58 2 22 
 
 400 
 
 65 6 16 
 
 357 
 
 5S 64 
 
 401 
 
 65 9 22 
 
 358 
 
 58 9 11 
 
 402 
 
 65 13 4 
 
 359 
 
 58 12 17 
 
 403 
 
 65 16 11 
 
 360 
 
 58 16 
 
 404 
 
 65 19 17 
 
 361 
 
 58 19 6 
 
 405 
 
 66 3 
 
 362 
 
 59 9 12 
 
 406 
 
 66 6 6 
 
 363 
 
 59 5 19 
 
 407 
 
 66 9 12 
 
 364 
 
 59 9 1 
 
 408 
 
 66 12 19 
 
 365 
 
 59 12 8 
 
 409 
 
 66 16 1 
 
 366 
 
 59 15 14 
 
 410 
 
 66 19 8 
 
 367 
 
 59 18 20 
 
 411 
 
 67 2 14 
 
 368 
 
 60 2 3 
 
 4T2 
 
 67 5 20 
 
ASSAY TABLE. 
 
 XXXIX 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 Ore give of 
 
 will yield of 
 
 Ore give of 
 
 will yield of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 413 
 
 67 9 3 
 
 457 
 
 74 12 20 
 
 414 
 
 67 12 9 
 
 458 
 
 74 16 3 
 
 415 
 
 67 15 16 
 
 459 
 
 74 19 9 
 
 416 
 
 67 18 22 
 
 460 
 
 75 2 16 
 
 417 
 
 68 2 4 
 
 461 
 
 75 5 22 
 
 418 
 
 68 5 11 
 
 462 
 
 75 9 4 
 
 419 
 
 68 8 17 
 
 463 
 
 75 12 11 
 
 420 
 
 68 12 
 
 464 
 
 75 15 17 
 
 421 
 
 68 15 6 
 
 465 
 
 75 19 
 
 422 
 
 68 18 12 
 
 466 
 
 76 2 6 
 
 423 
 
 69 1 19 
 
 467 
 
 76 5 12 
 
 424 
 
 69 5 1 
 
 468 
 
 76 8 19 
 
 425 
 
 69 8 8 
 
 469 
 
 76 12 1 
 
 426 
 
 69 11 14 
 
 470 
 
 76 15 8 
 
 427 
 
 69 14 20 
 
 471 
 
 76 18 14 
 
 428 
 
 69 18 3 
 
 472 
 
 77 1 20 
 
 429 
 
 70 1 9 
 
 473 
 
 77 5 3 
 
 430 
 
 70 4 16 
 
 474 
 
 77 8 9' 
 
 431 
 
 70 7 22 
 
 475 
 
 77 11 16 
 
 432 
 
 70 11 4 
 
 476 
 
 77 14 22 
 
 433 
 
 70 14 11 
 
 477 
 
 77 18 4 
 
 434 
 
 70 17 17 
 
 478 
 
 78 1 11 
 
 435 
 
 71 1 
 
 479 
 
 78 4 17 
 
 436 
 
 71 4 6 
 
 -480 
 
 78 8 
 
 437 
 
 71 7 12 
 
 481 
 
 78 11 6 
 
 438 
 
 71 10 19 
 
 482 
 
 78 14 12 
 
 439 
 
 71 14 1 
 
 483 
 
 78 17 19 
 
 440 
 
 71 17 8 
 
 484 
 
 79 1 1 
 
 441 
 
 72 14 
 
 485 
 
 79 4 8 
 
 442 
 
 72 3 20 
 
 486 
 
 79 7 14 
 
 443 
 
 72 7 3 
 
 487 
 
 79 10 20 
 
 444 
 
 72 10 9 
 
 488 
 
 79 14 3 
 
 445 
 
 72 13 16 
 
 489 
 
 79 17 9 
 
 446 
 
 72 16 22 
 
 490 
 
 80 16 
 
 447 
 
 73 4 
 
 491 
 
 80 3 22 
 
 448 
 
 73 3 11 
 
 492 
 
 80 7 4 
 
 449 
 
 73 6 17 
 
 493 
 
 80 10 11 
 
 450 
 
 73 10 
 
 494 
 
 80 13 17 
 
 451 
 
 73 13 6 
 
 495 
 
 80 17 
 
 452 
 
 73 16 12 
 
 496 
 
 81 6 
 
 453 
 
 73 19 19 
 
 497 
 
 81 3 12 
 
 454 
 
 74 3 1 
 
 498 
 
 81 6 19 
 
 455 
 
 74 6 8 
 
 499 
 
 81 10 1 
 
 45(5 
 
 74 9 14 
 
 500 
 
 81 13 8 
 
xl 
 
 ASSAY TABLE. 
 
 If 200 Grains of 
 
 One 
 
 Ton of Ore 
 
 If 200 Grains of 
 
 One 
 
 Ton of Ore 
 
 Ore give of 
 
 wi 
 
 11 yield of 
 
 Ore give of 
 
 will yield of 
 
 FINK METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 501 
 
 81 
 
 16 
 
 14 
 
 545 
 
 89 
 
 
 
 8 
 
 502 
 
 81 
 
 19 
 
 20 
 
 546 
 
 89 
 
 3 
 
 14 
 
 503 
 
 82 
 
 3 
 
 3 
 
 547 
 
 89 
 
 6 
 
 20 
 
 504 
 
 82 
 
 6 
 
 9 
 
 548 
 
 89 
 
 10 
 
 3 
 
 505 
 
 82 
 
 9 
 
 16 
 
 549 
 
 89 
 
 13 
 
 9 
 
 506 
 
 82 
 
 12 
 
 22 
 
 550 
 
 89 
 
 16 
 
 16 
 
 507 
 
 82 
 
 16 
 
 4 
 
 551 
 
 89 
 
 19 
 
 22 
 
 5C8 
 
 82 
 
 19 
 
 11 
 
 552 
 
 90 
 
 3 
 
 4 
 
 509 
 
 83 
 
 2 
 
 17 
 
 553 
 
 90 
 
 6 
 
 11 
 
 510 
 
 83 
 
 6 
 
 
 
 554 
 
 90 
 
 9 
 
 17 
 
 511 
 
 83 
 
 9 
 
 6 
 
 555 
 
 90 
 
 13 
 
 
 
 512 
 
 83 
 
 12 
 
 12 
 
 556 
 
 90 
 
 16 
 
 6 
 
 513 
 
 83 
 
 15 
 
 19 
 
 557 
 
 90 
 
 19 
 
 12 
 
 514 
 
 83 
 
 19 
 
 1 
 
 558 
 
 91 
 
 2 
 
 19 
 
 515 
 
 84 
 
 2 
 
 8 
 
 559 
 
 91 
 
 6 
 
 1 
 
 516 
 
 84 
 
 5 
 
 14 
 
 560 
 
 91 
 
 9 
 
 8 
 
 517 
 
 84 
 
 8 
 
 20 
 
 561 
 
 91 
 
 12 
 
 14 
 
 518 
 
 84 
 
 12 
 
 3 
 
 562 
 
 91 
 
 15 
 
 20 
 
 519 
 
 84 
 
 15 
 
 9 
 
 563 
 
 91 
 
 19 
 
 3 
 
 520 
 
 84 
 
 18 
 
 16 
 
 564 
 
 93 
 
 2 
 
 9 
 
 521 
 
 85 
 
 1 
 
 22 
 
 565 
 
 92 
 
 5 
 
 16 
 
 522 
 
 85 
 
 5 
 
 4 
 
 566 
 
 92 
 
 8 
 
 22 
 
 523 
 
 85 
 
 8 
 
 11 
 
 .567 
 
 92 
 
 12 
 
 4 
 
 524 
 
 85 
 
 11 
 
 17 
 
 568 
 
 92 
 
 15 
 
 11 
 
 525 
 
 85 
 
 15 
 
 
 
 569 
 
 92 
 
 18 
 
 17 
 
 526 
 
 85 
 
 18 
 
 6 
 
 570 
 
 93 
 
 2 
 
 
 
 527 
 
 86 
 
 1 
 
 12 
 
 571 
 
 93 
 
 5 
 
 6 
 
 528 
 
 86 
 
 4 
 
 19 
 
 572 
 
 93 
 
 8 
 
 12 
 
 529 
 
 86 
 
 8 
 
 1 
 
 573 
 
 93 
 
 11 
 
 19 
 
 530 
 
 86 
 
 11 
 
 8 
 
 574 
 
 93 
 
 15 
 
 1 
 
 531 
 
 86 
 
 14 
 
 14 
 
 575 
 
 93 
 
 18 
 
 8 
 
 532 
 
 86 
 
 17 
 
 20 
 
 576 
 
 94 
 
 1 
 
 14 
 
 533 
 
 87 
 
 1 
 
 3 
 
 577 
 
 94 
 
 4 
 
 20 
 
 534 
 
 87 
 
 4 
 
 9 
 
 578 
 
 94 
 
 8 
 
 3 
 
 535 
 
 87 
 
 7 
 
 16 
 
 579 
 
 94 
 
 11 
 
 9 
 
 536 
 
 87 
 
 10 
 
 22 
 
 580 
 
 94 
 
 14 
 
 16 
 
 537 
 
 87 
 
 14 
 
 4 
 
 581 
 
 94 
 
 17 
 
 22 
 
 538 
 
 87 
 
 17 
 
 11 
 
 582 
 
 95 
 
 1 
 
 4 
 
 539 
 
 88 
 
 
 
 17 
 
 583 
 
 95 
 
 4 
 
 11 
 
 540 
 
 88 
 
 4 
 
 
 
 584 
 
 95 
 
 7 
 
 17 
 
 541 
 
 88 
 
 7 
 
 6 
 
 585 
 
 95 
 
 11 
 
 
 
 542 
 
 88 
 
 10 
 
 12 
 
 586 
 
 95 
 
 14 
 
 6 
 
 543 
 
 88 
 
 13 
 
 19 
 
 587 
 
 95 
 
 17 
 
 12 
 
 544 
 
 88 
 
 17 
 
 1 
 
 588 
 
 96 
 
 
 
 19 
 
ASSAY TABLE. 
 
 xli 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 Ore give of 
 
 will yield of 
 
 Ore give of 
 
 will yield of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 Gr. 
 
 Oz. Dwts. Grs. 
 
 589 
 
 96 4 1 
 
 633 
 
 103 7 19 
 
 590 
 
 96 7 8 
 
 634 
 
 103 11 1 
 
 591 
 
 96 10 14 
 
 635 
 
 103 14 8 
 
 592 
 
 96 13 20 
 
 636 
 
 103 17 14 
 
 593 
 
 96 17 3 
 
 637 
 
 104 20 
 
 594 
 
 97 9 
 
 638 
 
 104 4 3 
 
 595 
 
 97 3 16 
 
 639 
 
 104 7 9 
 
 596 
 
 97 6 22 
 
 640 
 
 104 10 16 
 
 597 
 
 97 10 4 
 
 641 
 
 104 13 22 
 
 598 
 
 97 13 11 
 
 642 
 
 104 17 4 
 
 599 
 
 97 16 17 
 
 643 
 
 105 11 
 
 600 
 
 98 
 
 644 
 
 105 3 17 
 
 601 
 
 98 3 6 
 
 645 
 
 105 7 
 
 602 
 
 98 6 12 
 
 646 
 
 105 10 6 
 
 603 
 
 98 9 ]9 
 
 647 
 
 105 13 12 
 
 604 
 
 98 13 
 
 648 
 
 105 16 19 
 
 605 
 
 98 16 8 
 
 649 
 
 106 1 
 
 606 
 
 98 19 14 
 
 650 
 
 106 3 8 
 
 607 
 
 99 2 20 
 
 651 
 
 106 6 14 
 
 608 
 
 99 6 3 
 
 652 
 
 106 9 20 
 
 609 
 
 99 9 9 
 
 653 
 
 106 13 3 
 
 610 
 
 99 12 16 
 
 654 
 
 106 16 9 
 
 611 
 
 99 15 22 
 
 655 
 
 106 19 16 
 
 612 
 
 99 19 4 
 
 656 
 
 107 2 22 
 
 613 
 
 100 2 11 
 
 657 
 
 107 6 4 
 
 614 
 
 100 5 17 
 
 658 
 
 107 9 11 
 
 615 
 
 100 9 
 
 659 
 
 107 12 17 
 
 616 
 
 100 12 6 
 
 660 
 
 107 16 
 
 617 
 
 100 15 12 
 
 661 
 
 107 19 6 
 
 618 
 
 100 18 19 
 
 662 
 
 108 2 12 
 
 619 
 
 101 2 1 
 
 663 
 
 108 5 19 
 
 620 
 
 101 5 8 
 
 664 
 
 108 9 1 
 
 621 
 
 101 8 14 
 
 -665 
 
 108 12 8 
 
 622 
 
 101 11 20 
 
 666 
 
 108 15 14 
 
 623 
 
 101 15 3 
 
 667 
 
 108 18 20 
 
 624 
 
 101 18 9 
 
 668 
 
 109 2 3 
 
 625 
 
 102 1 16 
 
 669 
 
 109 5 9 
 
 626 
 
 102 4 22 
 
 670 
 
 109 8 16 
 
 627 
 
 102 8 4 
 
 671 
 
 109 11 22 
 
 628 
 
 102 11 11 
 
 672 
 
 109 15 4 
 
 629 
 
 102 14 17 
 
 673 
 
 109 18 11 
 
 630 
 
 102 18 
 
 674 
 
 110 1 17 
 
 631 
 
 103 1 6 
 
 675 
 
 110 5 
 
 632 
 
 103 4 12 
 
 676 
 
 110 8 6 
 
xlii 
 
 ASSAY TABLE. 
 
 If 200 Grains of 
 
 One 
 
 Ton of Ore 
 
 If 200 Grains 
 
 of One 
 
 Ton of Ore 
 
 Ore give of 
 
 will yield 
 
 of 
 
 Ore give of 
 
 will yield 
 
 of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 Gr. 
 
 o*: 
 
 Dwts. 
 
 Gra. 
 
 677 
 
 110 
 
 11 
 
 12 
 
 721 
 
 117 
 
 15 
 
 6 
 
 678 
 
 110 
 
 14 
 
 19 
 
 722 
 
 117 
 
 18 
 
 12 
 
 679 
 
 110 
 
 18 
 
 1 
 
 723 
 
 118 
 
 1 
 
 19 
 
 680 
 
 111 
 
 1 
 
 8 
 
 724 
 
 118 
 
 5 
 
 1 
 
 681 
 
 111 
 
 4 
 
 14 
 
 725 
 
 118 
 
 8 
 
 8 
 
 682 
 
 111 
 
 7 
 
 20 
 
 726 
 
 118 
 
 11 
 
 14 
 
 683 
 
 111 
 
 11 
 
 3 
 
 727 
 
 118 
 
 14 
 
 20 
 
 684 
 
 111 
 
 14 
 
 9 
 
 728 
 
 118 
 
 18 
 
 3 
 
 685 
 
 111 
 
 17 
 
 6 
 
 729 
 
 119 
 
 1 
 
 9 
 
 686 
 
 112 
 
 
 
 22 
 
 730 
 
 119 
 
 4 
 
 16 
 
 687 
 
 112 
 
 4 
 
 4 
 
 731 
 
 119 
 
 7 
 
 22 
 
 688 
 
 112 
 
 7 
 
 11 
 
 732 
 
 119 
 
 11 
 
 4 
 
 689 
 
 112 
 
 10 
 
 17 
 
 733 
 
 119 
 
 14 
 
 11 
 
 690 
 
 112 
 
 14 
 
 
 
 734 
 
 119 
 
 17 
 
 17 
 
 691 
 
 112 
 
 17 
 
 6 
 
 735 
 
 120 
 
 1 
 
 
 
 692 
 
 113 
 
 
 
 12 
 
 736 
 
 120 
 
 4 
 
 6 
 
 693 
 
 113 
 
 3 
 
 19 
 
 737 
 
 120 
 
 7 
 
 12 
 
 694 
 
 113 
 
 7 
 
 1 
 
 738 
 
 120 
 
 10 
 
 19 
 
 695 
 
 113 
 
 10 
 
 8 
 
 739 
 
 120 
 
 14 
 
 1 
 
 696 
 
 113 
 
 13 
 
 14 
 
 740 
 
 120 
 
 17 
 
 o 
 
 697 
 
 113 
 
 16 
 
 20 
 
 741 
 
 121 
 
 
 
 14 
 
 698 
 
 114 
 
 
 
 3 
 
 742 
 
 121 
 
 3 
 
 20 
 
 699 
 
 114 
 
 3 
 
 9 
 
 743 
 
 121 
 
 7 
 
 3 
 
 700 
 
 114 
 
 6 
 
 16 
 
 744 
 
 121 
 
 10 
 
 9 
 
 701 
 
 114 
 
 9 
 
 22 
 
 745 
 
 121 
 
 13 
 
 6 
 
 702 
 
 114 
 
 13 
 
 4 
 
 746 
 
 121 
 
 16 
 
 22 
 
 703 
 
 114 
 
 16 
 
 12 
 
 747 
 
 122 
 
 
 
 4 
 
 704 
 
 114 
 
 19 
 
 17 
 
 748 
 
 122 
 
 3 
 
 11 
 
 705 
 
 115 
 
 3 
 
 
 
 749 
 
 122 
 
 6 
 
 17 
 
 706 
 
 115 
 
 6 
 
 6 
 
 750 
 
 122 
 
 10 
 
 
 
 707 
 
 115 
 
 9 
 
 12 
 
 751 
 
 122 
 
 13 
 
 16 
 
 708 
 
 115 
 
 12 
 
 19 
 
 752 
 
 122 
 
 16 
 
 12 
 
 709 
 
 115 
 
 16 
 
 1 
 
 753 
 
 122 
 
 19 
 
 19 
 
 710 
 
 115 
 
 19 
 
 8 
 
 754 
 
 123 
 
 3 
 
 1 
 
 711 
 
 116 
 
 2 
 
 14 
 
 755 
 
 123 
 
 6 
 
 8 
 
 712 
 
 116 
 
 5 
 
 20 
 
 756 
 
 123 
 
 9 
 
 14 
 
 713 
 
 116 
 
 9 
 
 3 
 
 757 
 
 123 
 
 12 
 
 20 
 
 714 
 
 116 
 
 12 
 
 9 
 
 758 
 
 123 
 
 16 
 
 3 
 
 715 
 
 116 
 
 15 
 
 16 
 
 759 
 
 123 
 
 19 
 
 9 
 
 716 
 
 116 
 
 18 
 
 22 
 
 760 
 
 124 
 
 2 
 
 16 
 
 717 
 
 117 
 
 2 
 
 4 
 
 , -761 
 
 124 
 
 5 
 
 22 
 
 718 
 
 117 
 
 5 
 
 11 
 
 762 
 
 124 
 
 9 
 
 4 
 
 719 
 
 117 
 
 8 
 
 17 
 
 763 
 
 124 
 
 12 
 
 11 
 
 720 
 
 117 
 
 12 
 
 
 
 764 
 
 124 
 
 15 
 
 17 
 
ASSAY TABLE. 
 
 xliii 
 
 If 200 Grains 
 
 of One Ton of Ore 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 Ore give of 
 
 will 
 
 yield 
 
 of 
 
 Ore give of 
 
 will yield 
 
 of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. 
 
 Lwts. 
 
 Grs. 
 
 Gr. 
 
 0*. 
 
 Dwts. 
 
 Grs. 
 
 765 
 
 124 
 
 19 
 
 
 
 809 
 
 132 
 
 2 
 
 17 
 
 766 
 
 125 
 
 2 
 
 6 
 
 810 
 
 132 
 
 6 
 
 
 
 767 
 
 125 
 
 5 
 
 12 
 
 811 
 
 132 
 
 9 
 
 6 
 
 768 
 
 125 
 
 8 
 
 19 
 
 812 
 
 132 
 
 12 
 
 12 
 
 769 
 
 125 
 
 12 
 
 1 
 
 813 
 
 132 
 
 15 
 
 19 
 
 770 
 
 125 
 
 15 
 
 8 
 
 814 
 
 132 
 
 19 
 
 1 
 
 771 
 
 125 
 
 18 
 
 14 
 
 815 
 
 133 
 
 2 
 
 8 
 
 772 
 
 126 
 
 1 
 
 20 
 
 816 
 
 133 
 
 5 
 
 14 
 
 773 
 
 126 
 
 5 
 
 3 
 
 817 
 
 133 
 
 8 
 
 20 
 
 774 
 
 126 
 
 8 
 
 9 
 
 818 
 
 133 
 
 12 
 
 3 
 
 775 
 
 126 
 
 11 
 
 16 
 
 819 
 
 133 
 
 15 
 
 9 
 
 776 
 
 126 
 
 14 
 
 22 
 
 820 
 
 133 
 
 18 
 
 16 
 
 777 
 
 126 
 
 18 
 
 4 
 
 821 
 
 134 
 
 1 
 
 22 
 
 778 
 
 127 
 
 1 
 
 11 
 
 822 
 
 134 
 
 5 
 
 4 
 
 779 
 
 127 
 
 4 
 
 17 
 
 823 
 
 134 
 
 8 
 
 11 
 
 780 
 
 127 
 
 8 
 
 
 
 824 
 
 134' 
 
 1! 
 
 17 
 
 781 
 
 127 
 
 11 
 
 6 
 
 825 
 
 134 
 
 15 
 
 
 
 782 
 
 127 
 
 14 
 
 12 
 
 826 
 
 134 
 
 18 
 
 6 
 
 783 
 
 127 
 
 17 
 
 19 
 
 827 
 
 135 
 
 1 
 
 12 
 
 784 
 
 128 
 
 1 
 
 1 
 
 828 
 
 135 
 
 4 
 
 19 
 
 785 
 
 128 
 
 4 
 
 8 
 
 829 
 
 135 
 
 8 
 
 1 
 
 786 
 
 128 
 
 7 
 
 14 
 
 830 
 
 135 
 
 11 
 
 8 
 
 787 
 
 128 
 
 10 
 
 20 
 
 831 
 
 135 
 
 14 
 
 14 
 
 788 
 
 128 
 
 14 
 
 3 
 
 832 
 
 135 
 
 11 
 
 8 
 
 789 
 
 128 
 
 17 
 
 9 
 
 833 
 
 136 
 
 1 
 
 3 
 
 790 
 
 129 
 
 
 
 16 
 
 834 
 
 136 
 
 4 
 
 9 
 
 791 
 
 129 
 
 3 
 
 22 
 
 835 
 
 136 
 
 7 
 
 16 
 
 792 
 
 129 
 
 7 
 
 4 
 
 836 
 
 136 
 
 10 
 
 22 
 
 793 
 
 129 
 
 10 
 
 11 
 
 837 
 
 136 
 
 14 
 
 4. 
 
 794 
 
 129 
 
 13 
 
 17 
 
 838 
 
 136 
 
 17 
 
 11 
 
 795 
 
 129 
 
 17 
 
 
 
 839 
 
 137 
 
 
 
 17 
 
 796 
 
 130 
 
 
 
 6 
 
 840 
 
 137 
 
 4 
 
 
 
 797 
 
 130 
 
 3 
 
 12 
 
 841 
 
 137 
 
 7 
 
 6 
 
 798 
 
 130 
 
 6 
 
 19 
 
 842 
 
 137 
 
 10 
 
 12 
 
 799 
 
 130 
 
 10 
 
 1 
 
 843 
 
 137 
 
 13 
 
 19 
 
 800 
 
 130 
 
 13 
 
 8 
 
 844 
 
 137 
 
 17 
 
 1 
 
 801 
 
 130 
 
 16 
 
 14 
 
 845 
 
 138 
 
 
 
 8 
 
 802 
 
 130 
 
 19 
 
 20 
 
 846 
 
 138 
 
 3 
 
 14 
 
 803 
 
 131 
 
 3 
 
 3 
 
 847 
 
 138 
 
 6 
 
 20 
 
 804 
 
 131 
 
 6 
 
 9 
 
 848 
 
 138 
 
 10 
 
 3 
 
 805 
 
 131 
 
 9 
 
 16 
 
 849 
 
 138 
 
 13 
 
 19 
 
 806 
 
 131 
 
 12 
 
 22 
 
 850 
 
 138 
 
 16 
 
 16 
 
 807 
 
 131 
 
 16 
 
 4 
 
 851 
 
 138 
 
 19 
 
 22 
 
 808 
 
 131 
 
 19 
 
 11 
 
 852 
 
 139 
 
 3 
 
 4 
 
xliv 
 
 ASSAY TABLE. 
 
 If 200 Grains of 
 
 One 
 
 Ton of Ore 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 Ore give of 
 
 will yield 
 
 of 
 
 Ore give of 
 
 will 
 
 yield 
 
 of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 853 
 
 139 
 
 6 
 
 11 
 
 897 
 
 146 
 
 10 
 
 4 
 
 854 
 
 139 
 
 9 
 
 17 
 
 898 
 
 146 
 
 13 
 
 11 
 
 855 
 
 139 
 
 13 
 
 
 
 899 
 
 146 
 
 16 
 
 17 
 
 856 
 
 139 
 
 16 
 
 6 
 
 900 
 
 147 
 
 
 
 
 
 857 
 
 139 
 
 19 
 
 12 
 
 901 
 
 147 
 
 3 
 
 6 
 
 858 
 
 140 
 
 2 
 
 19 
 
 902 
 
 147 
 
 6 
 
 12 
 
 859 
 
 140 
 
 6 
 
 1 
 
 903 
 
 147 
 
 9 
 
 19 
 
 860 
 
 140 
 
 9 
 
 8 
 
 904 
 
 147 
 
 13 
 
 1 
 
 861 
 
 140 
 
 12 
 
 14 
 
 905 
 
 147 
 
 16 
 
 8 
 
 862 
 
 140 
 
 15 
 
 20 
 
 906 
 
 147 
 
 19 
 
 14 
 
 863 
 
 140 
 
 19 
 
 3 
 
 907 
 
 148 
 
 2 
 
 2 
 
 864 
 
 141 
 
 2 
 
 9 
 
 908 
 
 148 
 
 6 
 
 3 
 
 865 
 
 141 
 
 5 
 
 16 
 
 909 
 
 148 
 
 9 
 
 9 
 
 866 
 
 141 
 
 8 
 
 22 
 
 910 
 
 148 
 
 12 
 
 16 
 
 867 
 
 141 
 
 12 
 
 4 
 
 911 
 
 148 
 
 15 
 
 21 
 
 868 
 
 141 
 
 15 
 
 11 
 
 912 
 
 148 
 
 19 
 
 4 
 
 869 
 
 141 
 
 18 
 
 17 
 
 913 
 
 149 
 
 2 
 
 11 
 
 870 
 
 142 
 
 2 
 
 
 
 914 
 
 149 
 
 5 
 
 17 
 
 871 
 
 142 
 
 5 
 
 6 
 
 915 
 
 149 
 
 9 
 
 
 
 872 
 
 142 
 
 8 
 
 12 
 
 916 
 
 149 
 
 12 
 
 6 
 
 873 
 
 142 
 
 11 
 
 19 
 
 917 
 
 149 
 
 15 
 
 12 
 
 874 
 
 142 
 
 15 
 
 1 
 
 918 
 
 149 
 
 18 
 
 19 
 
 875 
 
 142 
 
 18 
 
 8 
 
 919 
 
 150 
 
 2 
 
 
 876 
 
 143 
 
 1 
 
 14 
 
 920 
 
 150 
 
 5 
 
 8 
 
 877 
 
 143 
 
 4 
 
 20 
 
 921 
 
 150 
 
 8 
 
 14 
 
 878 
 
 143 
 
 8 
 
 3 
 
 922 
 
 150 
 
 11 
 
 20 
 
 879 
 
 143 
 
 11 
 
 9 
 
 923 
 
 150 
 
 15 
 
 3 
 
 880 
 
 143 
 
 14 
 
 16 
 
 924 
 
 150 
 
 18 
 
 9 
 
 881 
 
 143 
 
 17 
 
 22 
 
 925 
 
 151 
 
 1 
 
 16 
 
 882 
 
 144 
 
 1 
 
 4 
 
 926 
 
 151 
 
 4 
 
 22 
 
 883 
 
 144 
 
 4 
 
 11 
 
 927 
 
 151 
 
 8 
 
 4 
 
 884 
 
 144 
 
 7 
 
 17 
 
 928 
 
 151 
 
 11 
 
 11 
 
 885 
 
 144 
 
 11 
 
 
 
 929 
 
 151 
 
 14 
 
 17 
 
 886 
 
 144 
 
 14 
 
 6 
 
 930 
 
 151 
 
 18 
 
 
 
 887 
 
 144 
 
 17 
 
 12 
 
 931 
 
 152 
 
 1 
 
 6 
 
 888 
 
 145 
 
 
 
 19 
 
 932 
 
 152 
 
 4 
 
 12 
 
 889 
 
 145 
 
 4 
 
 1 
 
 933 
 
 152 
 
 7 
 
 19 
 
 890 
 
 145 
 
 7 
 
 8 
 
 934 
 
 152 
 
 11 
 
 1 
 
 891 
 
 145 
 
 10 
 
 14 
 
 935 
 
 152 
 
 14 
 
 8 
 
 892 
 
 145 
 
 13 
 
 20 
 
 936 
 
 152 
 
 17 
 
 14 
 
 893 
 
 145 
 
 17 
 
 3 
 
 937 
 
 153 
 
 
 
 20 
 
 894 
 
 146 
 
 
 
 9 
 
 938 
 
 153 
 
 4 
 
 3 
 
 895 
 
 146 
 
 3 
 
 16 
 
 939 
 
 153 
 
 7 
 
 9 
 
 896 
 
 146 
 
 6 
 
 22 
 
 940 
 
 153 
 
 10 
 
 16 
 
ASSAY TABLE. 
 
 xlv 
 
 If 200 Grains of 
 
 One Ton of Ore 
 
 If 200 Grains 
 
 of One 
 
 Ton of Ore 
 
 Ore give of 
 
 will yield of 
 
 Ore give of 
 
 will yield of 
 
 FINE METAI, 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 GTS. 
 
 
 Gr. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 941 
 
 153 
 
 13 
 
 22 
 
 
 985 
 
 160 
 
 17 
 
 6 
 
 942 
 
 153 
 
 17 
 
 4 
 
 
 986 
 
 161 
 
 
 
 22 
 
 943 
 
 154 
 
 
 
 11 
 
 
 987 
 
 161 
 
 4 
 
 4 
 
 944 
 
 154 
 
 3 
 
 17 
 
 
 988 
 
 161 
 
 7 
 
 11 
 
 945 
 
 154 
 
 7 
 
 
 
 
 989 
 
 161 
 
 10 
 
 17 
 
 946 
 
 154 
 
 10 
 
 6 
 
 
 990 
 
 161 
 
 14 
 
 
 
 947 
 
 154 
 
 13 
 
 12 
 
 
 991 
 
 161 
 
 17 
 
 6 
 
 948 
 
 154 
 
 16 
 
 19 
 
 
 992 
 
 162 
 
 
 
 12 
 
 949 
 
 155 
 
 
 
 1 
 
 
 993 
 
 162 
 
 3 
 
 19 
 
 950 
 
 155 
 
 3 
 
 8 
 
 
 994 
 
 162 
 
 7 
 
 1 
 
 951 
 
 155 
 
 6 
 
 14 
 
 
 995 
 
 162 
 
 10 
 
 8 
 
 952 
 
 155 
 
 9 
 
 20 
 
 
 996 
 
 162 
 
 13 
 
 14 
 
 953 
 
 155 
 
 13 
 
 3 
 
 
 997 
 
 162 
 
 16 
 
 20 
 
 954 
 
 155 
 
 16 
 
 9 
 
 
 998 
 
 163 
 
 
 
 3 
 
 955 
 
 155 
 
 19 
 
 16 
 
 
 999 
 
 163 
 
 3 
 
 9 
 
 956 
 
 156 
 
 2 
 
 22 
 
 1 
 
 grain 
 
 163 
 
 6 
 
 16 
 
 957 
 
 156 
 
 6 
 
 4 
 
 2 
 
 
 326 
 
 13 
 
 8 
 
 958 
 
 156 
 
 9 
 
 11 
 
 3 
 
 
 490 
 
 
 
 
 
 959 
 
 156 
 
 12 
 
 17 
 
 4 
 
 
 653 
 
 6 
 
 16 
 
 960 
 
 156 
 
 16 
 
 
 
 5 
 
 
 816 
 
 13 
 
 8 
 
 961 
 
 156 
 
 19 
 
 6 
 
 6 
 
 
 980 
 
 
 
 
 
 962 
 
 157 
 
 2 
 
 12 
 
 7 
 
 
 1143 
 
 6 
 
 16 
 
 963 
 
 157 
 
 5 
 
 19 
 
 8 
 
 
 1306 
 
 13 
 
 8 
 
 964 
 
 157 
 
 9 
 
 1 
 
 9 
 
 
 1470 
 
 
 
 
 
 965 
 
 157 
 
 12 
 
 8 
 
 10 
 
 
 1633 
 
 6 
 
 16 
 
 966 
 
 157 
 
 15 
 
 14 
 
 11 
 
 
 1796 
 
 13 
 
 8 
 
 967 
 
 157 
 
 18 
 
 20 
 
 12 
 
 
 1960 
 
 
 
 
 
 968 
 
 158 
 
 2 
 
 3 
 
 13 
 
 
 2123 
 
 6 
 
 16 
 
 969 
 
 158 
 
 5 
 
 9 
 
 14 
 
 
 2286 
 
 13 
 
 8 
 
 970 
 
 158 
 
 8 
 
 16 
 
 15 
 
 
 2450 
 
 
 
 
 
 971 
 
 158 
 
 11 
 
 22 
 
 16 
 
 
 2613 
 
 6 
 
 16 
 
 972 
 
 158 
 
 15 
 
 4 
 
 17 
 
 
 2776 
 
 13 
 
 8 
 
 973 
 
 158 
 
 18 
 
 11 
 
 18 
 
 
 2940 
 
 
 
 
 
 974 
 
 159 
 
 1 
 
 17 
 
 19 
 
 
 3103 
 
 6 
 
 16 
 
 975 
 
 159 
 
 5 
 
 
 
 20 
 
 
 3266 
 
 13 
 
 8 
 
 976 
 
 159 
 
 8 
 
 6 
 
 21 
 
 
 3430 
 
 
 
 
 
 977 
 
 159 
 
 11 
 
 12 
 
 22 
 
 
 3593 
 
 6 
 
 16 
 
 978 
 
 159 
 
 14 
 
 19 
 
 23 
 
 
 3756 
 
 13 
 
 8 
 
 979 
 
 159 
 
 18 
 
 1 
 
 24 
 
 
 3920 
 
 
 
 
 
 980 
 
 160 
 
 1 
 
 8 
 
 25 
 
 
 4083 
 
 6 
 
 16 
 
 981 
 
 160 
 
 4 
 
 14 
 
 26 
 
 
 4246 
 
 13 
 
 8 
 
 982 
 
 160 
 
 7 
 
 20 
 
 27 
 
 
 4410 
 
 
 
 
 
 983 
 
 160 
 
 10 
 
 3 
 
 28 
 
 
 4573 
 
 6 
 
 16 
 
 984 
 
 160 
 
 14 
 
 9 
 
 29 
 
 
 4736 
 
 13 
 
 8 
 
xlvi 
 
 ASSAY TABLE. 
 
 If 200 Grains of One Ton of Ore 
 
 If 200 Grains 
 
 of One 
 
 Ton of Ore 
 
 Ore give 
 
 of will 
 
 yield of 
 
 Ore give of will yield of 
 
 FINE METAL FINE METAL 
 
 FINE METAL FINE METAL 
 
 Grs. 
 
 0*: 
 
 Dwts. 
 
 Grs. 
 
 Grs. 
 
 Oz. 
 
 Dwts. 
 
 Grs. 
 
 30 
 
 4900 
 
 
 
 
 
 74 
 
 12086 
 
 13 
 
 8 
 
 31 
 
 5063 
 
 6 
 
 16 
 
 75 
 
 12250 
 
 
 
 
 
 32 
 
 5226 
 
 13 
 
 8 
 
 76 
 
 12413 
 
 6 
 
 16 
 
 33 
 
 5390 
 
 
 
 
 
 77 
 
 12576 
 
 13 
 
 8 
 
 34 
 
 5553 
 
 6 
 
 16 
 
 78 
 
 12740 
 
 
 
 
 
 35 
 
 5716 
 
 13 
 
 8 
 
 79 
 
 12903 
 
 6 
 
 16 
 
 36 
 
 5880 
 
 
 
 
 
 80 
 
 13066 
 
 13 
 
 8 
 
 37 
 
 6043 
 
 6 
 
 16 
 
 81 
 
 13230 
 
 
 
 
 
 38 
 
 6206 
 
 13 
 
 8 
 
 82 
 
 13393 
 
 6 
 
 16 
 
 39 
 
 6370 
 
 
 
 
 
 83 
 
 13556 
 
 13 
 
 8 
 
 40 
 
 6533 
 
 6 
 
 16 
 
 84 
 
 13720 
 
 
 
 
 
 41 
 
 6696 
 
 13 
 
 8 
 
 85 
 
 13883 
 
 6 
 
 16 
 
 42 
 
 6860 
 
 
 
 
 
 86 
 
 14046 
 
 13 
 
 8 
 
 43 
 
 7023 
 
 6 
 
 16 
 
 87 
 
 14210 
 
 
 
 
 
 44 
 
 7186 
 
 13 
 
 8 
 
 88 
 
 14373 
 
 6 
 
 16 
 
 45 
 
 7350 
 
 
 
 
 
 89 
 
 14536 
 
 13 
 
 8 
 
 46 
 
 7513 
 
 6 
 
 16 
 
 90 
 
 14700 
 
 
 
 
 
 47 
 
 7676 
 
 13 
 
 8 
 
 91 
 
 14863 
 
 6 
 
 16 
 
 48 
 
 7840 
 
 
 
 
 
 92 
 
 15026 
 
 13 
 
 8 
 
 49 
 
 8003 
 
 6 
 
 16 
 
 93 
 
 15190 
 
 
 
 
 
 50 
 
 8166 
 
 13 
 
 8 
 
 94 
 
 15353 
 
 6 
 
 16 
 
 51 
 
 8330 
 
 
 
 
 
 95 
 
 15516 
 
 13 
 
 8 
 
 52 
 
 8493 
 
 6 
 
 16 
 
 96 
 
 15680 
 
 
 
 
 
 53 
 
 8656 
 
 13 
 
 8 
 
 97 
 
 15843 
 
 6 
 
 16 
 
 54 
 
 8820 
 
 
 
 
 
 98 
 
 16006 
 
 13 
 
 8 
 
 55 
 
 8983 
 
 6 
 
 16 
 
 99 
 
 16170 
 
 
 
 
 
 56 
 
 9146 
 
 13 
 
 8 
 
 100 
 
 16333 
 
 6 
 
 16 
 
 57 
 
 9310 
 
 
 
 
 
 101 
 
 16496 
 
 13 
 
 8 
 
 58 
 
 9473 
 
 6 
 
 16 
 
 102 
 
 16660 
 
 
 
 
 
 59 
 
 9636 
 
 13 
 
 8 
 
 103 
 
 16823 
 
 6 
 
 16 
 
 60 
 
 9800 
 
 
 
 
 
 104 
 
 16986 
 
 13 
 
 8 
 
 61 
 
 9963 
 
 6 
 
 16 
 
 105 
 
 17150 
 
 
 
 
 
 62 
 
 10126 
 
 13 
 
 8 
 
 106 
 
 17313 
 
 6 
 
 16 
 
 63 
 
 10290 
 
 
 
 
 
 107 
 
 17476 
 
 13 
 
 8 
 
 64 
 
 10453 
 
 6 
 
 16 
 
 108 
 
 17640 
 
 
 
 
 
 65 
 
 10616 
 
 13 
 
 8 
 
 109 
 
 17803 
 
 6 
 
 16 
 
 66 
 
 10780 
 
 
 
 
 
 110 
 
 17966 
 
 13 
 
 8 
 
 67 
 
 10943 
 
 6 
 
 16 
 
 111 
 
 18130 
 
 
 
 
 
 68 
 
 11106 
 
 13 
 
 8 
 
 112 
 
 18293 
 
 6 
 
 16 
 
 69 
 
 11270 
 
 
 
 
 
 113 
 
 18456 
 
 13 
 
 8 
 
 70 
 
 11433 
 
 6 
 
 16 
 
 114 
 
 18620 
 
 
 
 
 
 71 
 
 11596 
 
 13 
 
 8 
 
 115 
 
 18783 
 
 6 
 
 16 
 
 72 
 
 11760 
 
 
 
 
 
 116 
 
 18946 
 
 13 
 
 8 
 
 73 
 
 11923 
 
 6 
 
 16 
 
 117 
 
 19110 
 
 
 
 
 
ASSAY TABLE. 
 
 xlvii 
 
 If 200 Grains 
 
 of One 
 
 Ton of Ore 
 
 If 200 Grains of One 
 
 Ton of Ore 
 
 Ore give of 
 
 will yield 
 
 of 
 
 Ore give of 
 
 will yield 
 
 of 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 FINE METAL 
 
 Grs. 
 
 a*. 
 
 Dwts 
 
 Grs. 
 
 Grs. 
 
 02. 
 
 Dwts 
 
 Grs. 
 
 118 
 
 19273 
 
 6 
 
 16 
 
 160 
 
 26133 
 
 6 
 
 16 
 
 119 
 
 19436 
 
 13 
 
 8 
 
 161 
 
 26296 
 
 13 
 
 8 
 
 120 
 
 19600 
 
 
 
 
 
 162 
 
 26460 
 
 
 
 
 
 121 
 
 19763 
 
 6 
 
 16 
 
 163 
 
 26623 
 
 6 
 
 16 
 
 122 
 
 19926 
 
 13 
 
 8 
 
 164 
 
 26786 
 
 13 
 
 8 
 
 123 
 
 20090 
 
 
 
 
 
 165 
 
 26950 
 
 
 
 
 
 124 
 
 20253 
 
 6 
 
 16 
 
 166 
 
 27113 
 
 6 
 
 16 
 
 125 
 
 20416 
 
 13 
 
 8 
 
 167 
 
 27276 
 
 13 
 
 8 
 
 126 
 
 20580 
 
 
 
 
 
 168 
 
 27440 
 
 
 
 
 
 127 
 
 20743 
 
 6 
 
 16 
 
 169 
 
 27603 
 
 6 
 
 16 
 
 128 
 
 20906 
 
 13 
 
 8 
 
 170 
 
 27766 
 
 13 
 
 8 
 
 129 
 
 21070 
 
 
 
 
 
 171 
 
 27930 
 
 
 
 
 
 130 
 
 21233 
 
 6 
 
 16 
 
 172 
 
 28093 
 
 6 
 
 16 
 
 131 
 
 21396 
 
 13 
 
 8 
 
 173 
 
 28256 
 
 13 
 
 8 
 
 132 
 
 21560 
 
 
 
 
 
 174 
 
 28420 
 
 
 
 
 
 133 
 
 21723 
 
 6 
 
 16 
 
 175 
 
 28583 
 
 6 
 
 16 
 
 134 
 
 21886 
 
 13 
 
 8 
 
 176 
 
 28746 
 
 13 
 
 8 
 
 135 
 
 22050 
 
 
 
 
 
 177 
 
 28910 
 
 
 
 
 
 136 
 
 22213 
 
 6 
 
 16 
 
 178 
 
 29073 
 
 6 
 
 16 
 
 137 
 
 22376 
 
 13 
 
 8 
 
 179 
 
 29236 
 
 13 
 
 8 
 
 138 
 
 22540 
 
 
 
 
 
 180 
 
 29400 
 
 
 
 
 
 139 
 
 22703 
 
 6 
 
 16 
 
 181 
 
 29563 
 
 6 
 
 16 
 
 140 
 
 22866 
 
 13 
 
 8 
 
 182 
 
 29726 
 
 13 
 
 8 
 
 141 
 
 23030 
 
 
 
 
 
 183 
 
 29890 
 
 
 
 
 
 142 
 
 23193 
 
 6 
 
 16 
 
 184 
 
 30053 
 
 6 
 
 16 
 
 143 
 
 23356 
 
 13 
 
 8 
 
 185 
 
 30216 
 
 13 
 
 8 
 
 144 
 
 23520 
 
 
 
 
 
 186 
 
 30380 
 
 
 
 
 
 . 145 
 
 23683 
 
 6 
 
 16 
 
 187 
 
 30543 
 
 6 
 
 16 
 
 146 
 
 23846 
 
 13 
 
 8 
 
 188 
 
 3U706 
 
 13 
 
 8 
 
 147 
 
 24010 
 
 
 
 
 
 189 
 
 30870 
 
 
 
 
 
 148 
 
 24173 
 
 6 
 
 16 
 
 190 
 
 31033 
 
 6 
 
 16 
 
 149 
 
 24336 
 
 13 
 
 8 
 
 191 
 
 31196 
 
 13 
 
 8 
 
 150 
 
 24500 
 
 
 
 
 
 192 
 
 31360 
 
 
 
 
 
 151 
 
 24663 
 
 6 
 
 16 
 
 193 
 
 31523 
 
 6 
 
 16 
 
 152 
 
 24826 
 
 13 
 
 8 
 
 194 
 
 31686 
 
 13 
 
 8 
 
 153 
 
 24990 
 
 
 
 
 
 195 
 
 31850 
 
 
 
 
 
 154 
 
 25153 
 
 6 
 
 16 
 
 196 
 
 32013 
 
 6 
 
 16 
 
 155 
 
 25316 
 
 13 
 
 8 
 
 197 
 
 32176 
 
 13 
 
 8 
 
 156 
 
 25480 
 
 
 
 
 
 198 
 
 32340 
 
 
 
 
 
 157 
 
 25643 
 
 6 
 
 16 
 
 199 
 
 32503 
 
 6 
 
 16 
 
 158 
 
 25806 
 
 13 
 
 8 
 
 200 
 
 32666 
 
 13 
 
 8 
 
 159 
 
 25970 
 
 
 
 
 
 
 
 
 
INDEX. 
 
 ABS 
 
 ABSOLUTE heating power of fuel, 
 154 
 Acid, boracic, 221 
 
 coloured flame, 295 
 
 nitric, 213 
 
 oxalic, 174 
 
 phosphoric, coloured flame of, 296 
 
 tartaric, 174 
 Acids, 3 
 
 Addison's process for the estimation of 
 
 carbon in steel, 367 
 Adhesive paste, 115 
 Agate, 250 
 
 mortars, 20* 
 Alabaster, 261 
 
 Alkalies, action of, on galena, 503 
 
 caustic, 181 
 
 in iron ores, 355 
 Alkaline persulphides, 190 
 
 Alloys, metallic, capable of direct cupel- 
 
 lation, 729, 730 
 incapable of direct cupellation, 
 
 729, 733 
 
 of antimony and lead, assay of, 565 
 copper, 434 
 
 - gold, 740, 741, 768, 769 
 
 silver, platinum, and copper, 
 
 759 
 
 standard of, 768 
 
 lead, assay of, 530 
 
 silver, 600 
 
 and copper, assay 1 of, 627, 632 
 
 - zinc, assay of, 574 
 Almandine, 884 
 
 ruby, 257 
 Almond meal, 114 
 Alum, 288 
 
 Alumina crucibles, 130 
 in iron ores, 341, 349 
 Aluminium blowpipe support, 212 
 - silicate, 192 
 
 silicates, 291 
 Amalgam, assay of, 737 
 Amalgamation of silver, 626 
 Amethyst, 249, 890 
 Ammonium fluoride, 222 
 Analysis, colorimetric, 307 
 
 ARS 
 
 Analysis of fuel, 150 
 
 iron ores, calculation of results, 
 
 349 
 platinum ores, 800 
 
 volumetric, 298 
 Anglesite, 271, 289 
 Anthracite, 171 
 Antimonial silver, assay of, 735 
 
 or arsenical silver ores, 282 
 
 silver ore, 284, 286 
 
 ores, 276 
 
 Antimonides of silver, 633 
 Antimonium crudum, assay of, 557 
 Antimoniuretted from arseniuretted 
 
 hydrogen, to distinguish, 564 
 Antimony and arsenic, separation of tin 
 from, 564 
 
 and lead, assay of alloys of, 565 
 sulphides, 277 
 
 assay of, 556 
 
 coloured flaine, 295 
 
 in' iron ores, 356 
 
 sublimates, detection of, 563 
 
 regulus, assay of, 558 
 
 separation of, from bismuth, 815 
 
 sulphide, 186, 189, 263, 284 
 assay of, 557 
 
 sulphides, 282 
 Anvil, 14 
 
 Apatite, 258, 289, 290 
 
 Apparatus for silver cupellation before 
 
 the blowpipe, 717 
 Appendix, i. 
 Aquamarine, 254, 892 
 Aqua regia, parting with, 766 
 Aragonite, 260, 288 
 Argentiferous tin, assay of, 734 
 Argol, 199 
 
 action of, on galena, 503 
 
 reducing power of, 607 
 Armentiferous zinc, assay of, 735 
 Arnold's apparatus for analysis of iron 
 
 and steel, 373 
 
 process for estimation of chromium 
 in iron and steel, 827 
 
 Arquerite, assay of, 737 
 Arseniate of copper, 434 
 
 3p 
 
INDEX. 
 
 ARS 
 
 Arseniate of iron, 287 
 Arsenic, 264, 284 
 
 - and antimony, separation of tin 
 from, 564 
 
 assay of, 831 
 
 colour of flame, 295 
 
 in copper, 499 
 iron ores, 356 
 
 separation of, from bismuth, 814 
 
 sulphide, 264, 286 
 
 sulphur, nickel and cobalt, assay 
 of, 851 
 
 Arsenical nickel, 284 
 -268 
 
 or antimonial silver ores, 282 
 
 pyrites, 265 
 
 silver ore, 284, 286 
 ores, 276 
 
 Arseniuretted from antimoniuretted 
 
 hydrogen, to distinguish, 564 
 Asbestos, 252 
 
 card, 213 
 Ash in fuel, 70 
 estimation of, 162 
 
 pit, 58 
 
 Assay balances, 28 
 
 before blowpipe, 235 
 
 complete, of iron ores, 332 
 
 crucible, of silver ores, 605 
 
 English copper, 435 
 
 furnace, gas, 87 
 
 in scorifier, 615 
 
 Lake Superior copper, 455 
 
 mode of taking from ingot, 687 
 
 of antimonial silver, 735 
 
 antimony, 556 
 
 argentiferous iron and steel, 737 
 
 mercury, 737 
 tin, 734 
 
 arsenic, 831 
 
 auriferous copper ores, 775 
 bell metal, 734 
 
 bismuth, 809 
 
 brass, 733 
 
 bronze, 734 
 
 chromium, 820 
 
 copper, electrolytic, 480 
 
 in the wet way, 461 
 
 fuel, 150 
 
 galena, 521 
 
 gold, 740 
 
 coin and bullion, 769 
 gun metal, 734 
 
 iron, 308 
 
 in the dry way, 309 
 
 wet way, 321 
 
 lead, 502 
 
 - by fusion with black flux, 510 
 - iron, 511 
 
 potassium carbonate, 
 604 
 
 sodium carbonate or 
 black flux and metallic iron, 514 
 
 BER 
 
 Assay of lead, roasting and reducing, 
 515 
 
 with standard solutions, 531 
 sulphuric acid, 519 
 
 manganese, 833 
 
 mercury, 587 
 
 blowpipe, 598 
 
 nickel and cobalt, 839 
 
 platinum, 781 
 ores, 795 
 
 pure tin oxide, 539 
 
 pyrites for gold, 773 
 
 saltpetre, 177, 178 
 
 silver, 600 
 
 and copper, 739 
 
 before the blowpipe, 715 
 
 - by the wet way, 634, 647, 674 
 
 sulphur, 859 
 
 in wet way, 861 
 telluric silver, 735 
 
 zinc, 567 
 
 volumetrical, 575 
 
 ' pound,' 694 
 
 spectroscopic of gold, 779 
 
 table for gold, xxxiv 
 
 volumetric of bismuth, 817 
 
 chromium, 827 
 
 mercury, 595 
 Asterias, 256 
 
 Atkinson's modification of Penny's 
 volumetric assay of iron, 325 
 
 Atomic weights, 2, 3, 6 
 
 Attwood's blowpipe assay of mercury, 
 598 
 
 Aufrye and d'Arcet's muffle furnace, 63 
 
 Augite, 252, 289, 290 
 
 Azurite, 434 
 
 BALANCE, the, 26 
 theory of the, 28 
 Balas ruby, 257 
 Balling's process for titration of silver 
 
 in galena, 714 
 Bank value into mint value, table to 
 
 convert, xxxiii 
 
 Bankers or bullion balance, 27 
 Bars of furnace, 59 
 Baryta, coloured flame of, 296 
 Barytes, 262 
 
 Basic cinder in manufactured iron, 414 
 Beale's cement, 120 
 Bears from smelting furnaces, assay of 
 
 737 
 Becker's process for assay of antimony 
 
 563 
 
 Beech -wood ash, 143 
 Bell metal, assay of, 734 
 
 tin in, 549, 550 
 
 Berthier's assay of mercury, 590 
 
 method for assay of fuel, 155 
 Beryl, 254 
 
 blue, 889 
 
INDEX. 
 
 li 
 
 BER 
 
 Beryl or emerald, 293 
 Bettell's process for estimating cinder 
 in iron, 415 
 
 titanium, 432 
 Binary substances, 3, 5 
 Bioitite, 869 
 
 Bismuth, 263, 284, 286 
 
 assay of, 809 
 
 cupellation of silver in, 730 
 
 effect of, on ductility of silver, 699 
 
 ores, 809 
 
 purification of, from arsenic, 814 
 
 antimony, 815 
 
 copper, 815 
 sulphur, 816 
 
 refining crude, 813 
 
 volumetric, assay of, 817 
 Bisulphite of ammonium for reduction 
 
 of iron salts, 328 
 Black band iron stone, 308 
 
 cupric oxide, 274 
 
 flux, 197 
 
 assay of lead by fusion with, 510 
 
 lead, 258 
 
 crucibles. 123 
 
 or graphite, 170 
 
 Blast furnace, 60 
 cinder, 310 
 Blende, 187, 269, 285, 287, 289 
 
 assay of, 573 
 
 cupriferous assay of, 573 
 Bloodstone, 250 
 
 Blossom's process of assaying gold ores, 
 
 741 
 Blowpipe and its uses, 202 
 
 assay of coal,165 
 
 mercury, 598 
 
 nickel, 849 
 
 silver, 715 
 
 operations, 234 
 
 reagents and fluxes, 213 
 Blue copper carbonate, 275 
 
 flames, 294 
 
 Boeckmaiin's assay of sulphur, 866 
 
 Boiler cement, 120 
 
 Boiling or evaporating furnace, 110, 
 
 990 
 Bone-ash, 717 
 
 for cupels, 141 
 
 ashes, 224 
 
 Boracic acid, coloured flame, 295 
 
 - vitrified, 221 
 Borate of lead, 177, 201 
 Borax, 193, 218 
 
 bead, colour of, 246 
 Boron, test for, 222 
 Bournonite, 434 
 Brasque crucibles, 312 
 Brass, assay of, 584, 733 
 
 on glass cement, 114 
 Braunite, 833 
 Brazil-wood paper, 213 
 Brick muffle furnace, 64 
 
 CAR 
 
 Bricks, magnesia, 130 
 Brightening of silver, 620 
 Britton's burette, 305 
 
 modification of Eggertz's estimation 
 of combined carbon in iron and steel, 
 374 
 
 Bromine, coloured flame, 295 
 Bronze, assay of, 584, 734 
 
 tin in, 550 
 Bronzite, 869 
 Brown ochre, 267 
 Brown's gas assay furnace, 87 
 
 volumetric assay of copper, 479 
 Brunton's automatic sampling machine, 
 
 11 
 
 Bruyeres cement, 120 
 Buisson's volumetric assay of lead, 533 
 Bullion, assay of, 769 
 
 silver, assay of, 631 
 Bunsen's gas burner, 105, 113 
 
 process for assay of manganese ores, 
 834 
 
 platinum ores, 782 
 
 Burette, 302, 636, 637 
 
 Burse's process for assay of tin in 
 
 bronze, 550 
 Busteed's process for assay of silver in 
 
 Indian mints, 687 
 
 ,250 
 
 \J Calamine, 270, 289 
 Calcination, 42 
 Calcining furnace, 55 
 Calcite, 259 
 Calcium, coloured flame of, 297 
 
 fluoride, 195, 223 
 
 silicate, 192 
 
 sulphate, 223 
 Gale-spar, 259, 288 
 Calculation of results, 164 
 Caoutchouc, 117 
 
 Carbon as a reducing agent, 170 
 
 combined, in iron and steel, estima- 
 tion of, 371 
 
 in iron, inorganic standards for the 
 colorimetric test, 381 
 
 and steel, 363 
 
 estimation of minute 
 
 quantities of, 376 
 
 ores, 351 
 
 steels, estimation of, 367 
 
 Carbonaceous matter in iron ores, 351 
 Carbonate of copper, 287, 289 
 
 iron, 268 
 
 lead, 271 
 
 sodium, 214 
 
 zinc, 270 
 
 Carbonates, alkaline, 187 
 
 of copper, blue and green, 275 
 
 potassium and sodium, 181, 195 
 
 Carbonic acid in iron ores, 349 
 
 3 p 2 
 
Hi 
 
 INDEX. 
 
 CAR 
 
 Carbonisation of fuel, volatile products 
 
 of, 161 
 
 Carnelian, 250 
 Carnelly's colorimetric assay of copper, 
 
 470 
 
 Caustic alkalies and carbonates, 187 
 Cement for brass on glass, 114 
 mending pestles, 115 
 
 waterproof, 116 
 Cementation, 47 
 Ceruse, 201 
 
 or white lead, 176 
 Cerusite, 288, 271, 289 
 Chalcedony, 250 
 Chalcopyrite, 434 
 
 Chapman's process for detection of 
 antimony in sublimates, 563 
 
 detection of copper in iron pyrites, 
 497 
 
 Charcoal. 182 
 
 crucibles, 126 
 
 for blowpipe, 210 
 furnaces, 70 
 
 fuel, examination of 161, 182 
 
 lined crucibles, 312 
 
 Chemical characters of minerals, 244 
 Chiastolite, 291 
 Chimney, 56, 59 
 Chisel, cold, 16 
 Chloride of silver, 224, 277 
 reduction of, 684 
 sodium, 196 
 
 process for assay of silver, 687 
 Chlorine, blue flame of, 295 
 Chlorite, 251, 290 
 Chlorcspinel, 257 
 
 Chromate of copper, 434 
 Chrome diallage, 869 
 
 iron ore, assay of, 820 
 Chromic iron, 267, 285, 287 
 Chromium, assay of, 820 
 
 in iron and steel, estimation of, 827 
 ores, 317 
 
 Chromometer, new form of, 380 
 . Chrysoberyl, 257, 293, 879 
 Chrysolite, 252, 291, 891 
 Chrysoprase, 892 
 Cinnabar, 189, 272, 282, 284, 286, 287 
 
 in ore assay of, 592 
 Cinnamon stone, 886 
 
 Clarke's process for assay of chrome 
 
 iron ore, 824 
 Claus's method of decomposing osmi- 
 
 ridium, 803 
 Clay iron stone, 308 
 Cleaning platinum crucibles, 136, 138 
 Cloud's colorimetric assay of copper, 467 
 Coal, 258 
 
 assay of, before the blowpipe, 165 
 
 valuation of, for the production of 
 illuminating gas, 167 
 
 Coating retorts, 119 
 
 Cobalt and nickel, separating, 839, 846 
 
 COP 
 
 Cobalt assay of, 839 
 
 for traces of, in nickel, 849 
 
 bloom, 268, 287 
 
 glance assay of, 851 
 
 in iron ores, 344 
 
 nitrate, 221 
 
 ores, 839, 846 
 
 speiss, assay of, 843, 851 
 
 tin-white, 269 
 Coin, gold, assay of, 769 
 Cold chisel, 16 
 Colorimetric analysis, 307 
 
 copper assay, 461 
 Coloured flames, 294 
 Colour of borax bead, 246 
 
 minerals, 239 
 Coke, 171 
 
 for furnaces, 70 
 
 from fuel, examination of, 161 
 Compactness of fuel, 152 
 Condensation, 52 
 
 Copper, 213, 284 
 
 alloys, 434 
 
 and lead speiss, assay of, 855 
 silver, assay of, 739 
 
 alloys, assay of 627, 632 
 
 arseniate, 434 
 
 arsenic in, 499 
 
 assay colorimetric, 461 
 
 in the wet way, 461 
 
 Lake Superior, 455 
 
 of, 434 
 
 volumetric, 476 
 
 sulphide, argentiferous, 600 
 
 carbonate, 287 
 
 carbonates, blue and green, 275 
 
 chromate, 434 
 
 coloured flame of, 296 
 
 electrolytic, assay of, 480 
 
 from zinc, separation of, 584 
 
 glance, 434 
 
 gold alloys, assay of, 756 
 
 grey, 274, 282, 283, 284 
 
 in iron ores, 356 
 
 pyrites, 497 
 
 native, 273 
 
 nickel, 831. 
 
 ores, 434 
 
 assay of auriferous, 775 
 
 oxide, 181, 201, 221 
 
 red, 285 
 
 and black, 287 
 
 phosphate, 434 
 
 pyrites, 273, 282, 434 . 
 assay of, for sulphur, 859 
 
 494, 496 
 
 selenide, 277 
 
 silicate, 434 
 
 silver and platinum alloys, assay of, 
 633 
 
 alloys, cupellation of, 732 
 
 separation of, from bismuth, 815 
 
 sulphate, 181, 288 
 
INDEX. 
 
 liii 
 
 COP 
 
 Copper sulphide, 185 
 
 titration of silver in presence of, 
 713 
 
 vanadate, 434 
 
 vitreous, 282 
 
 zinc and nickel alloys, assay of, 853 
 Cornish crucibles, 120 
 
 Corundum, 256, 293 
 Coruscation of silver, 620 
 Covellite, 434 
 Cream of tartar, 199 
 Crucible assay of gold ores, 742 
 
 perforated, 361 
 Crucibles, cupels, &c., 120 
 
 for assay of iron ores, 311 
 calcination, 42 
 
 mounts for, 107 
 
 supporting, 66 
 Crystalline form, 237 
 Cube, 237 
 
 Cupel bottoms, assay of, 525 
 
 moulds, 142 
 
 or muffle furnace, 62 
 Cupels, 141, 618 
 
 crucibles, &c., 120 
 Cupellation, 54, 617, 720 
 
 - loss, 727 
 
 loss in assay of copper and silver 
 alloys, 630 
 
 of gold, 749 
 
 and silver, lead to be employed 
 
 in, 759 
 
 silver before the blowpipe, 716 
 
 in lead or bismuth, 730 
 
 Cupric oxide, black, 274 
 Cupriferous bismuth, assay of, 810 
 
 blende, assay of, 573 
 Cuprous oxide, red, 274, 285, 287 
 Cyanide of potassium, 214 
 Cyanite, 888 
 
 Cyanosite, 434 
 Cymophane, 879 
 
 DANIEL'S pyrometer, 144 
 D'Arcet and Aufrye's muffle fur- 
 nace, 63 
 
 Day standard colours, 382 
 
 Day's process for assay of chrome iron 
 ore, 825 
 
 Debray's process for assay of platinum 
 ores, 795 
 
 Decinormal solutions, 300 
 
 De Clauby's process for volumetric 
 assay of tin, 550 
 
 Desulphurising agents, 181 
 
 Deutecom's assay of sulphur, 864 
 
 Deville's furnace, 62 
 
 process for assay of platinum ores 
 795 
 
 tin in gun and bell 
 
 metal, 549 
 
 Diamond, 257, 292, 294, 867 
 
 FER 
 
 Dichroite, 889 
 Diehl's assay of lead, 534 
 )isthene, 888 
 Distillation, 51 
 
 of mercury, 587 
 
 zinc, 568 
 
 Distilled water, 52 
 
 Dodecahedron, 237 
 Dolomite, 260, 288 
 Domeykite, 434 
 
 Draught crucible gas furnace, 91 
 Dressing, washing, or vanning, 23 
 Drill for silver ingots, 687 
 
 [)rip proof gas burner, 112 
 
 Dry assay of iron, 309 
 
 distillation, 51 
 
 Dumonte's assay of lead with standard 
 
 solutions, 531 
 Dyce's process for separating silver from 
 
 base metals, 633 
 
 T7ARTHENWARB retorts, 53 
 
 Jj Effects produced by wind and 
 
 blast furnaces, 72 
 Egg and lime lute, 114 
 Eggertz's colour test for combined car- 
 bon in iron and steel, 372 
 
 process for estimating sulphur in 
 iron and steel, 388 
 
 the estimation of combined 
 
 carbon in iron and steel, 371 
 
 graphite in iron and 
 
 steel, 369 
 
 Electrolytic assay of copper, 480 
 
 mercury, 592 
 
 Elements, 1, 2, 3 
 
 Elutriation, 22 
 
 Emerald, 254, 292, 293 
 
 yellow, 881 
 
 green, 892 
 Emery, 256 
 
 Endemann's colorimetric assay of 
 
 copper, 467 
 
 English copper assay, 435 
 Epidote, 290 
 Epsomite, 288 
 Eschka's process for assay of mercury, 
 
 591 
 Escosura's [L. de la] electrolytic assay 
 
 of mercury, 592 
 Essonite, 886 
 Evaporating furnaces, 57 
 
 FjUHLERZ, 434 
 
 argentiferous, 600 
 
 Faraday's blast furnace, 60 
 
 blowpipe directions, 206 
 
 Fat lute, 113 
 Fatty oils, 172 
 Felspar, 255, 290, 895 
 Ferric oxide, 266 
 
liv 
 
 INDEX. 
 
 FER 
 
 Ferric oxide, brown and red, 287 
 
 in iron ores, 341 
 
 Ferro-manganese, manganese in, 426 
 
 Ferrous oxide in iron ores, 346, 347 
 
 Fire lute, 113 
 
 Fish-eye, 895 
 
 Flame for blowpipe, 205 
 
 Flasks, measuring, 307 
 
 Fleck's volumetric assay of copper, 476 
 
 Fleitmann's process for assay of traces 
 
 of cobalt in nickel. 849 
 Fletcher's draught crucible furnace, 91 
 
 drip proof gas burner, 112 
 
 gas reverberatory furnace, 90 
 injector gas furnace, 92 
 
 muffle gas furnace, 91 
 
 petroleum furnace, 94 
 
 safety Bunsen burner, 113 
 
 solid flame gas burner, 111 
 
 universal gas furnace, 84 
 Flint, 250 
 
 Fluoride of ammonium, 222 
 
 calcium, 223 
 
 Fluor spar, 195, 259, 290 
 Flux, Turner's, 222 
 Flux, white, black, and raw, 197 
 Fluxes, 191 
 
 blowpipe, 213 
 
 for iron ores, 309, 310 
 
 metallic, 201 
 
 reducing power of, 200 
 Fluxing crucibles, 122 
 
 Forbes's blowpipe assay of silver, 715 
 
 charcoal, 210 
 
 directions, 203 
 
 lamp, 206 
 
 charcoal borers, 211 
 
 glass and platinum forceps, 50 
 
 lime crucibles, 129 
 
 soda-paper, 233 
 
 Forceps, glass and platinum, 50 
 Fracture of fuel, 152 
 
 minerals, 238 
 
 Franklinite, 308 
 
 Fresenius's assay of copper pyrites, 496 
 
 Fuel, assay of, 150 
 
 external appearance of, 152 
 
 for furnaces, 70 
 Furnace, blast, 60 
 
 calcining, 55 
 
 evaporating, 57 
 
 fusion, 57 
 
 operations, 66 
 
 universal, 65 
 
 wind, 57 
 Furnaces, 55 
 Fusion, 48 
 Fusion furnace, 57 
 Fusibility of minerals, 244 
 
 GALENA, 189, 270, 282 
 action of oxygen on, 502 
 
 GOL 
 
 Galena action of metallic iron on, 502 
 alkalies and alkaline carbo- 
 nates on, 502 
 
 potassium nitrate on, 502 
 argol on, 502 
 
 assay of, in the wet way, 521 
 
 separation of silver from, 627 
 
 titration of silver in, 714 
 Galletti's process for assay of zinc, 575 
 Garnet, 253, 290, 291, 293, 884 
 
 noble, 886 
 Garrett's burette, 340 
 Gas assay furnace, 87 
 
 burner, Bunsen's, 105 
 solid flame, 111 
 
 furnaces, 74 
 
 illuminating, valuation of coal for 
 the production of, 167 
 
 reverberatory furnace, 90 
 Gay-Lussac's burette, 304 
 Gems, 291 
 
 and precious stones, discrimination 
 of, 867 
 
 Genth's process for assay of chrome 
 
 iron ore, 820 
 
 German silver, assay of, 733 
 Gibbs's process for assay of chrome iron 
 
 ore, 823 
 separating nickel and cobalt, 
 
 846 
 
 of assaying platinum ores, 804 
 
 Glanzkobalt, 831 
 Glass, 193, 290 
 
 analysis of various kinds of, 194 
 
 and artificial gems, 896 
 
 forceps, 50 
 
 of antimony, 557 
 lead, 201 
 
 retorts, 53 
 
 to brass, cementing, 114 
 Globules of silver, weight of, 723 
 Gooch's perforated crucible, 361 
 Gore's gas furnace, 95 
 
 Gozdorf's method of estimating, weight 
 
 of spheres of, 753 
 Graphite, 170, 258, 285 
 
 crucibles, 123 
 
 in iron and steel, estimation of, 369, 
 370 
 
 Green copper carbonate, 275 
 
 flames, 294 
 
 Grey copper, 274, 282, 283, 284 
 Greenearth, 267, 287 
 Griffin's oil furnace, 74 
 
 reverberatory gas furnace, 100 
 Grinding and powdering ores, 332 
 Gold, 275, 284 
 
 and palladium, rhodium, silver, and 1 
 mercury, 740 
 
 platinum, 799 
 
 alloys for pyrometry, 149 
 
 silver, melting by gas, 92 
 
 separating, 767 
 
INDEX. 
 
 Iv 
 
 GOL 
 
 Gold assay of pyrites for, 773 
 weights for, 35 
 
 assays, apparatus for boiling, 763 
 
 coin, assay of, 740, 769 
 
 copper, alloys, assay of, 756 
 
 cupellation of, 749 
 
 estimating weight of minute spheres 
 of, 753 
 
 in copper ores, assay of, 775 
 
 minerals, detection of traces of, 777 
 
 leaf (false), assay of, 584 
 
 silver, platinum, and copper alloys, 
 759 
 
 standard of alloys of, 768 
 
 valuing tables, ii, xx, xxi, xxxiii 
 Gum, 174 
 
 Gun metal, assay of, 734 
 
 tin in, 549, 550 
 Guyard's process for assay of platinum 
 
 residues, 799 
 Gypsum, 223, 261, 290 
 
 HADOW'S process for separation of 
 cobalt and nickel, 839 
 Hallet's process for assay of tin, 548 
 Hammers, 15 
 Hard cement, 117 
 Hardness of gems, 291 
 
 iron and steel, 433 
 
 minerals, 240 
 
 Haswell's process for titration of iron, 
 
 330 
 
 Hausmannite, 833 
 
 Heat, production and application of, 55 
 Heating power of fuel, estimation of, 
 
 154 
 
 Heavy spar, 262, 290 
 Heine's colorimetric copper assay, 461 
 Hematite, 266 
 
 red and brown, 308 
 
 Herpin's electrolytic assay of copper 
 
 and nickel, 853 
 Hessian crucibles, 120 
 Holland's assay of sulphur, 865 
 Hood, 57 
 
 Hornblende, 252, 289, 290 
 Horn silver, 277, 289 
 Hornstone, 250 
 
 Houzeau's assay of sulphur, 864 
 Hyacinth, 254, 884 
 Hydrogen gas, 169 
 
 reduction by, 48 
 Hyposulphite of sodium for titration of 
 
 iron, 330 
 
 TNCINERATING precipitates, 41 
 
 JL Ingot mould, 69 
 
 Injector gas furnace, 92 
 
 Inquartation, 761 
 
 Iodide of silver, 230 
 
 Iodine, coloured flame of, 296 
 
 tincture of, 229 
 
 KUN 
 
 Iridium, 782, 785 
 
 and platinum, 797 
 Iron, 182, 284 
 
 arseniate, 287 
 
 assay of, 308 
 
 assay of lead by fusion with, 511 
 
 and silver, assay of, 737 
 
 steel, assay of chromium in, 827 
 
 estimation of silicon in, 407, 
 413 
 
 hardness of, 433 
 
 estimation of sulphur in, 388 
 
 carbonate, 268, 289 
 
 cement, 119 
 
 chromic, 285, 287 
 
 crucibles, malleable, 133 
 
 dry, assay of, 309 
 
 estimation of carbon, sulphur, sili- 
 con, phosphorus, &c., in, 363 
 
 in ores, estimation of, 338 
 
 magnetic, 266, 285 
 
 manganese in, 422 
 
 manufactured, basic cinder and oxide 
 in, 414 
 
 metallic, as a reducing agent, 175 
 
 mortar, 17 
 
 native, 265 
 
 new colorimetrical process for esti- 
 mating sulphur in, 393 
 
 on galena, action of, 502 
 
 ores, estimation of sulphur in, 338, 
 405 
 
 list of, 308 
 
 manganese in, 426 
 
 phosphoric acid in, 334 
 
 oxides, 201 
 
 oxide, brown, 285 
 
 pyrites, assay of, for sulphur, 859 
 
 peroxide, 181 
 
 phosphate, 287 
 
 phosphorus in, 416 
 
 pyrites, 190, 265, 280 
 magnetic, 278, 28?- 
 
 retorts, 53 
 
 specular, 285 
 
 sulphate, 181, 288 
 
 sulphide, 183 
 
 titanic, 285 
 
 titanium in, 430, 432 
 
 wire, 213 
 
 TACQUELAIN'S colorimetric assay of 
 tl copper, 472 
 Jargon, 879 
 Jasper, 250 
 
 Juptner's (von) method of separating 
 gold and silver, 767 
 
 TTIMBERLITE, 870 
 
 JV. Klaproth's process for assay of tin, 
 
 546 
 Kunzel's process for assay of zinc, 577 
 
Ivi 
 
 INDEX. 
 
 LAD 
 
 T ABLE, wrought iron, 69 
 
 JJ Lake Superior copper assay, 455 
 
 Laws of combination, 1, 5 
 
 Lea's process for assay of platinum 
 
 ores, 790 
 Le Blay's colorimetric assay of copper, 
 
 467 
 Lead alloys, assay of, 530 
 
 and antimony, alloys of, assay of, 
 565 
 
 sulphides, 277, 282 
 
 assay of, 502 
 
 by fusion with black flux, 510 
 
 iron, 511 
 
 potassium carbonate, 
 504 
 
 sodium carbonate, or 
 
 black flux and metallic iron, 514 
 
 with standard solutions, 531 
 
 with sulphuric acid, 519 
 
 remarks on, 517 
 
 borate, 177, 201 
 
 carbonate, 271 
 assay of, 523 
 
 colour of flame, 295 
 
 cupellation of silver in, 730 
 
 fumes, assay of, 525 
 
 in iron ores, 356 
 
 metallic, as a reducing agent, 176 
 
 nitrate, 
 
 nitrate, 17, 188 
 
 ores, 502 
 
 phosphate, 271 
 
 proof, 224 
 
 roasting and reducing, assay of, 
 515 
 
 selenide, 277 
 
 silicate, 201 
 
 silicates, 177 
 
 slags, assay of, 525 
 
 speiss, assay of, 855 
 
 sulphate, 181, 188, 201, 271 
 assay of, 527 
 
 sulphide, 187, 271, 284 
 
 to be employed in cupellation of 
 gold and silver, 759 
 
 Lens pocket, 622 
 
 Lenssen's process for volumetric assay 
 
 of tin, 551 
 Lersen's process for separation of zinc 
 
 from copper, 584 
 Level's assay of lead with potassium 
 
 ^ferrocyanide and cyanide, 519 
 Liebig's condenser, 52 
 
 process for separating nickel and 
 cobalt, 846 
 
 Lime and egg lute, 114 
 
 and its silicate, 192 
 
 coloured flame of, 297 
 
 crucibles, 128 
 
 in iron ores, 341 
 Lined crucibles, 127 
 Linseed meal, 114 
 
 MET 
 
 Litharge, 176, 182, 201 
 
 action of, on sulphides, 183 
 
 assay of, for silver, 608 
 
 assay of, 523 
 
 silver assay with, 602 
 Lithia, coloured flame of, 296 
 Litmus paper, 213 
 
 London crucibles, 120 
 
 Lowe, Mr., apparatus for boiling gold 
 
 assays, 763 
 Luckow's electrolytic assay of copper, 
 
 487 
 
 Lustre of minerals, 239 
 Lutes and cements, 113 
 Luting vessels for furnace operations, 
 
 118 
 Lyte's assay of lead, 529 
 
 ll/TACKINTOSH on estimating phos- 
 ITJ_ phorus in iron and steel, 417 
 Magnesia crucibles and bricks, 130 
 
 in iron ores, 341 
 Magnesite, 260,289 
 Magnesium silicate, 192 
 Magnetic iron, 266, 285 
 
 pyrites, 279, 282 
 
 ore, 308 
 
 Malleable iron crucibles, 133 
 Manganese, assay of, 833 
 
 in iron, 422 
 
 ores, 317,341 
 
 speigleisen, 426 
 
 ores, 268, 285, 287 
 assay of, 833 
 
 peroxide, 180 
 
 sulphide, 183 
 Manganite, 833 
 
 Markus on lead assay, 517 
 Mascazzini's assay of lead, 529 
 Matrix of diamond, 868 
 Measuring flasks, 307 
 
 the heat of a furnace, 144 
 Mechanical treatment of iron ores, 332 
 Meerschaum, 251 
 
 Melting arrangement (gas) for gold 
 and silver, 92 
 
 points of metals, 147 
 Mercury and gold, 740 
 
 assay of, 587 
 
 argentiferous, 737 
 
 silver alloys containing, 686 
 
 blowpipe, assay of, 598 
 
 electrolytic, assay of, 592 
 
 native, 272 
 
 selenide, 277 
 
 sulphide, 272 
 
 titration of silver in presence of, 
 714 
 
 volumetric assay of, 595 
 
 Merrick's assay of pyrites for gold, 773 
 
 Metallic fluxes, 201 
 
INDEX. 
 
 Ivii 
 
 MET 
 
 Metallic iron as a reducing agent, 175 
 
 lead as a reducing agent, 176 
 Metals, melting points of, 147 
 Method of weighing, 36 
 
 Mica, 251, 255, 286, 290 
 Micacsous iron, 266 
 Microcosmic salt, 219 
 Millerite, 277 
 Minerals, determination of, 279 
 
 discrimination of, 237 
 Minium, assay of, 523 
 
 Mint value, to convert, into bank value, 
 
 xxxiii 
 
 Mispickel, 265, 283 
 Mohr's burette, 302 
 
 process for assay of manganese ores, 
 835 
 
 - zinc, 582 
 
 volumetric assay of copper, 476 
 Moissenet on the English copper assay, 
 
 435 
 
 process for assay of tin, 548 
 Molybdenite, 263, 282, 286 
 Monger's process for assay of cupri- 
 ferous blende, 573 
 
 Mortars, 17 
 Mould, ingot, 69 
 Mounts for crucibles, 107 
 Muffle gas furnace, 81, 91 
 
 or cupel furnace, 62 
 Muir's test for bismuth, 818 
 Mundic, 265 
 
 \TEPHRITE, 251 
 
 ll Nickel and cobalt, separating, 839, 
 846 
 
 arsenical, 268, 284 
 
 assay for traces of cobalt in, 849 
 of, 839 
 
 commercial assay of, 851 
 metallic assay of, 845 
 
 crucibles, 140 
 
 glance, assay of, 851 
 
 in iron ores, 344 
 
 ores, 839, 845 
 
 oxalate, 221 
 
 pyrites, assay of, 851 
 - sulphide, 277, 283 
 
 white, 278, 284 
 
 zinc, and copper alloys, assay of, 
 853 
 
 Nipples, blowpipe, 204 
 Nitrate of cobalt, 221 
 
 lead, 180, 188 
 
 potassium, 220 
 
 sodium, 180 
 
 Nitrates of potassium and sodium, 177 
 Nitre, 188, 196, 220, 261, 288 
 
 oxidising power of, 607 
 Nitric acid, 213 
 Nomenclature, chemical, 1 
 Normal solutions, 300 
 
 PER 
 
 OCTAHEDRON, 237 
 Ohl's assay of nickel speiss, 854 
 Oil and gas furnaces, 74 
 Oil of cassia, testing gems in, 873 
 Oils, fatty, 172 
 Olassen's assay of zinc, cobalt, and 
 
 nickel, 850 
 Olivine, 252, 869 
 O'Neill's process for assay of chrome 
 
 iron ore, 822 
 Onyx, 250 
 Opal, 250, 291 
 Ores of antimony, 556 
 
 arsenic, 831 
 
 gold, 740 
 lead, 502 
 
 manganese, 833 
 
 mercury, 587 
 
 platinum, 781 
 
 silver, 600 
 
 principal of copper, 434 
 Oriental topaz, 256 
 
 amethyst, 256 
 
 emerald, 256 
 
 ruby, 256 
 
 Osmiridium, 795, 796, 803 
 Oudemans's process for titration of 
 
 iron, 330 
 
 Oxalate of nickel, 221 
 Oxalic acid, 174 
 Oxidation before blowpipe, 209 
 Oxide of copper, 181, 221 
 
 red and black, 287 
 
 iron, brown and red, 287 
 
 iron in manufactured iron, 414 
 
 - tin, 537 
 , red cuprous, 285 
 Oxides, 3, 4 
 
 of copper and iron, 201 
 Oxidised ores of copper, 434 
 Oxidising agents, 176 
 Oxychloride of zinc cement, 120 
 Oxygen on galena, action of, 502 
 
 PALLADIUM, 782, 798 
 JL and gold, 740 
 
 titration of silver in presence of, 
 714 
 
 Parnell's process for assay of arsenic, 
 
 832 
 Parting assay, 771 
 
 of gold and silver, 761 
 Paste, adhesive, 115 
 
 Pattinson's process for assay of man- 
 ganese ores, 836 
 
 Paull's assay of manganese ores, 835 
 Pearson's assay of copper pyrites, 494 
 
 process for assay of sulphur, 861 
 volumetric assay of bismuth, 817 
 
 Penny's process for assay of iron in wet 
 
 way, 321 
 Peridot, 891 
 
Iviii 
 
 INDEX. 
 
 PER 
 
 Peridotite, 871 
 
 Perofskite, 869 
 
 Peroxide of iron, 181 
 
 Perry's process for assay of platinum 
 
 ores, 807 
 Personne's process for volumetric assay 
 
 of mercury, 595 
 Pestle and mortar, 17 
 Pestles, c., mending, 115 
 Peters on the Lake Superior copper 
 
 assay, 455 
 
 Petroleum furnace, 94 
 Phosphate of copper, 434 
 
 iron, 287 
 
 lead, 271 
 
 sodium and ammonium, 219 
 Phosphoric acid, coloured flame of, 296 
 
 estimation of, in iron ores, 334 
 
 Phosphorus in iron, 317 
 
 _ and steel, 363, 416 
 
 Pinchbeck, assay of, 584 
 Pipette, 306 
 
 Pitchblende, 278, 286, 287 
 Plaster of Paris, 114 
 Platinum, 275, 284 
 
 and gold alloys for pyrometry, 149 
 iridium, 797 
 
 silver, assay of alloys of, 632 
 
 assay of, 781 
 
 crucibles, 133 
 
 preservation of, 134 
 
 forceps, 50 
 
 ores, analysis of, 782, 790 
 
 silver and copper alloys, assay of, 633 
 gold, and copper alloys, 759 
 
 spoons, 212 
 
 wire for blowpipe, 211 
 Plattner's detection of nickel before 
 
 blowpipe, 849 
 Pleonast, 257 
 Pliers, 622 
 
 Plumbago crucibles, 123 
 Pokers, 67 
 Polybasite, 600 
 Porcelain crucibles, 122 
 
 mortar, 17 
 Porosity of fuel, 152 
 Potash, 181 
 Potassium binoxalate, 199 
 
 bisulphate, 221 
 
 bitartrate, 199 
 
 carbonate, 181, 195 
 
 coloured flame of, 297 
 
 cyanide, 214 
 
 nitrate, 177, 188, 196, 220 
 
 on galena, action of, 503 
 
 Pound assay,' 694 
 Precipitates, incinerating, 41 
 
 weighing moist, 39 
 
 Price on a source of error in sulphur 
 
 estimations, 860 
 Prism, 238 
 Proof lead, 224 
 
 ROS 
 
 Prospecting for gold, &c., 26 
 Psilomelane, 833 
 Pulverisation, 18 
 j Pyrites, arsenical, 265, 831 
 
 assay of, for gold, 773 
 
 sulphur, 859 
 
 copper, 273 
 
 assay of, 494, 496 
 
 iron, 190, 289 
 
 magnetic iron, 278, 282 
 Pyrolusite, 833 
 Pyrometers, 144 
 
 Pyrometric heating power of fuel, 154 
 
 power of fuel, 160 
 Pyromorphite, 271, 288, 289 
 Pyrope, 869 
 
 AUARTATION, 761 
 U Quartz, 291, 292, 874 
 Quartz and silicates, 248 
 -violet, 890 
 
 yellow, 884 
 Quicksilver, 272 
 
 T)AMMELSBERG'S process for assay 
 
 XL of tin ores, 554 
 
 Raw flux, 197 
 
 Realgar, 831 
 
 Red copper, Malachite, 434 
 
 cuprous oxide, 274 
 
 flames, 295 
 
 ochre, 266 
 Reduction, 46 
 
 - before blowpipe, 209, 216 
 of iron salts by stannous chloride, 
 327 
 
 sulphite of ammonium, 
 
 328 
 
 zinc, 327 
 
 Reducing agents, 169 
 
 power of various agents, 175 
 
 fluxes, 200 
 
 Regulus of antimony, assay of, 558 
 
 Resins, 172 
 
 Resinous or hard cement, 117 
 
 Retorts, 53 
 
 Reverberatory furnace gas, 90, 100 
 
 Rhodium, 782, 785, 799 
 
 and gold, 740 
 Rhombohedron, 238 
 
 Riley's manganese in speigleisen, 424 
 
 process for estimating titanium in 
 iron, 430 
 
 Roasting, 44 
 
 gold ores, 745 
 Roasting-test, 44 
 Rock crystal, 249 
 
 salt, 261 
 Roman cement, 114 
 Rose quartz, 250 
 
 Rose's method of separating gold and 
 silver, 767 
 
INDEX. 
 
 lix 
 
 ROS 
 
 Boss's aluminium support, 212 
 Kubicelle, 257 
 Kuby, 293 
 
 silver, 600 
 
 spinel, 886 
 
 Rumpf's and Scherer's process for 
 
 assay of manganese ores, 834 
 Ruthenium, 782, 785 
 Rutile, 278, 287 
 
 SALAMANDER ' brand of crucibles, 
 121 
 
 Salt, 196 
 - common, 288 
 
 normal solution of, temperature cor- 
 rection, 651 
 
 of sorrel, 199 
 
 standard solution of, 636, 641 
 Salts, 3, 4 
 
 Saltpetre, 177, 188 
 Sample, preparation of the, 8 
 Sampling machine, 11 
 automatic, 11 
 
 steel and iron borings, 364 
 Sapphire, 256, 293 
 
 white, 876 
 
 yellow, 879 
 
 red, 886 
 - blue, 889 
 
 violet, 890 
 
 green, 891 
 
 water, 889 
 Sard, 250 
 Sardonyx, 250 
 
 Scale of hardness of minerals, 240 
 Schaffner's process for assay of zinc, 
 
 577 
 Scherer and Rumpf 's process for assay 
 
 of manganese ores, 834 
 Schwarz's process for assay of zinc, 581 
 Schwartz's volumetric assay of lead, 532 
 Scorification, 54, 610 
 
 assay of gold, 747 
 Scorifier, assay in, 615 
 Scorifiers, 144 
 
 Scully, effect of bismuth on ductility of 
 
 silver, 699 
 
 Sef Strom's blast furnace, 61 
 Selenides of mercury, lead, silver, and 
 
 copper, 277 
 Selenite, 261 
 Selenium, 280 
 
 coloured name, 295 
 
 Sell's process for volumetric assay of 
 
 chromium, 829 
 Serpentine, 251, 290 
 Sexton on arsenic in copper, 499 
 Shears, 17 
 Shimer's process for assay of metallic 
 
 iron and steel, 363 
 Siemens's pyrometer, 148 
 Sieve, 20 
 
 SIZ 
 
 Sifting, 22 
 Silica, 192, 222 
 
 in iron ores, 341 
 Silicate of copper, 434 
 
 lead, 177, 201 
 
 zinc, 279, 289 
 
 assay of, 572 
 
 Silicates of aluminium, 291 
 
 calcium, magnesium, aluminium, 
 
 192 
 Silicon, estimation of, in iron and steel, 
 
 407, 413 
 
 in iron and steel, 363 
 Silver, 276, 284 
 
 a blowpipe assay of, 715 
 
 alloys containing mercury, assay of, 
 686 
 
 and copper, assay of, 739 
 
 alloys, assay of, 627, 632 
 
 gold separating, 767 
 
 iron, assay of, 737 
 
 mercury, assay, of, 737 
 
 platinum, assay of alloys of, 632 
 
 antimonial, assay of, 735 
 
 assay of, 600 
 
 - by the wet way, 634, 647, 674 
 
 litharge for, 608 
 
 weights for, 35 
 
 with litharge, 602 
 
 bullion, assay of, 631 
 
 chloride, 77, 224, 
 reduction of, 684 
 
 copper alloys, cupellation of, 732 
 
 crucibles, 140 
 
 effect of bismuth on ductility of, 699 1 
 
 from galena, separation of, 627 
 
 glance, 600 
 
 gold, platinum, and copper alloys,, 
 759 
 
 horn, 289 
 
 in lead or bismuth, cupellation of,. 
 730 
 
 iodide, 231 
 
 lead, concentration of, 717 
 
 native, assay of, 633 
 
 ore, antimonial, 284 
 arsenical, 284 
 
 ores, arsenical and antimonial, 276 
 or antimonial, 282 
 
 ore, arsenical and antimonial, 286 
 
 ores, 600 
 
 platinum, and copper alloys, assay 
 of, 633 
 
 preparation of pure, 685 
 
 selenide, 277 
 
 separation of, from base metals, 633 
 
 sulphide, 276, 283 
 
 telluric, 735 
 
 titration of, in presence of, 714 
 
 copper, 713 
 
 with ammonium sulphocyanide,. 
 
 711 
 Size and shape of fuel, 152 
 
Ix 
 
 INDEX. 
 
 SKE 
 
 Skey's method of detecting traces of 
 
 gold in minerals, 777 
 Skittle pots, 120 
 Slag, iron, 310 
 Smaltine, 269, 284 
 Smaragdite, 869 
 Smith, J. Lawrence, on phosphorus in 
 
 iron and steel, 419 
 Smithsonite, 270 
 Smoky quartz, 250 
 Soap, 199 
 Soda, 181 
 
 coloured flame of, 296 
 - paper (Forbes's), 233 
 Sodium ammonio-phosphate, 219 
 
 biborate, 218 
 
 bisulphate, 220 
 
 carbonate, 181, 195, 214 
 
 chloride, 196 
 
 coloured flame of, 297 
 
 nitrate, 177, 180 
 
 sulphate, 181 
 Soft cement, 114 
 Solution, 49 
 Sonstadt's solution, 242 
 
 for testing gems, 873 
 
 Spathic iron ore, 308 
 Specific gravity of fuel, 153 
 
 iron ores, 360 
 
 minerals, 241 
 
 heating power of fuel, 160 
 Spectacles, neutral tint, 68 
 Spectroscopic assay of gold, 779 
 Specular iron, 266, 285 
 Spiegeleisen, manganese in, 424, 426 
 Speiskobalt, 831 
 
 Sphene, 278, 291 
 
 Spheres of gold, estimating weight of, 
 
 753 
 Spinel, 256, 292, 294 
 
 ruby, 886 
 Spoon platinum, 212 
 Srubescite, 434 
 Standard colours, day, 382 
 night, 386 
 
 of alloys of gold, 768 
 
 solution of silver, 640 
 
 solutions, 302, 300 
 
 Standards, inorganic, for colorimetric 
 carbon test, 381 
 
 Stannous chloride, reduction of iron 
 salts by, 327 
 
 Starch, 173 
 
 : Stead on the estimation of minute quan- 
 tities of carbon in iron and steel, 
 376 
 
 Steel and iron, estimation of silicon in, 
 407, 413 
 
 argentiferous assay of, 737 
 
 estimation of carbon, sulphur, sili- 
 con, phosphorus, &c., in, 363 
 
 sulphur in, 388 
 
 hardness of, 433 
 
 SUT 
 
 Steel mortar, 19, 332 
 
 phosphorus in, 416 
 
 Steinbeck's electrolytic assay of copper, 
 
 480 
 Stromeyer's process for volumetric assay 
 
 of tin, 551, 554 
 Streak of minerals, 239 
 Stones, precious discrimination of, 867 
 Storer's assay of galena, 521 
 Stourbridge clay crucibles, 122 
 Strontia, coloured flame of, 296 
 Sublimates, detection of antimony in, 
 
 563 
 
 Sublimation, 54 
 Sugar, 173 
 Sulphate of calcium, 223 
 
 lead, 188, 201, 271 
 
 sodium, 181 
 
 Sulphates of lead, copper, and iron, 
 
 181 
 zinc, iron, and copper, 288 
 
 soluble, 261 
 
 Sulphide of antimony, 189, 263, 284 
 
 assay of, 557 
 
 arsenic, 264, 286 
 
 copper, argentiferous, 600 
 
 lead, 271 
 
 mercury, 272 
 
 nickel, 277, 283 
 
 silver, 276 
 
 zinc, 285 
 
 assay of, 573 
 Sulphides, action of litharge on, 183 
 
 alkaline, 190 
 
 lead and antimony, 282 
 
 of lead and antimony, 277 
 Sulphite of ammonium for reduction of 
 
 iron salts, 328 
 Sulphocyanide of ammonium, titration 
 
 of silver with, 711 
 Sulphur, 189, 262, 286 
 
 arsenic, cobalt, and nickel assay of, 
 851 
 
 assay of, 859 
 in wet way, 861 
 
 in fuel, estimation of, 162 
 iron, 317 
 
 and steel, 363 
 
 estimation of, 388 
 
 new colorimetrical process for 
 
 estimating, 393 
 ores, estimation of, 338, 405 
 
 ores, 859 
 
 , separation of, from bismuth, 816 
 Sulphuric acid, assay of lead with, 519 
 
 in iron ores, 348 
 
 Sulphurising agents, 189 
 Sulphuretted ores of copper, 434 
 Sulphurous earth, 859 
 Supporting crucibles in gas furnace, 80 
 Supports, blowpipe, 210 
 Button's process for assay of antimony, 
 563 
 
INDEX. 
 
 SYM 
 
 Symbols, chemical, their employment 
 and uses, 7 
 
 STABLE, assay for gold, xxxiv 
 JL of cupellation loss, 728 
 the corrections of the standard 
 
 solutions of common salt, 654 
 weights of colourless stones in 
 
 air and water, 878 
 
 yellow stones in air and 
 water, 883 
 
 brown, 885 
 
 - red, 887 
 
 - blue, 890 
 
 violet, 891 
 
 green, 893 
 
 chatoyant, 895 
 for conversion of mint 
 
 value into bank value, xxxiii 
 
 Tables of quantity of fine gold in alloys 
 and the mint value, ii 
 
 Talbott's process for assay of tin in 
 presence of tungsten, 546 
 
 Talc, 250, 290 
 
 Tallow, 172 
 
 Tamm's process for assay of bismuth 
 ores, 809 
 
 Tantin's process for estimating phos- 
 phorus in iron and steel, 416 
 
 Tartaric acid, 174 
 
 Telluric silver, assay of, 735 
 
 Tellurium, coloured flame of, 296 
 
 graphic and foliated, assay of, 740 
 Temperature correction for normal 
 
 solution of salt, 651 
 
 Tennantite, 831 
 
 Terreil's process for separating nickel 
 and cobalt, 846 
 
 Testing minerals, requirements for, 
 248 
 
 Tetrahedron, 237 
 
 Thompson's, Dr., scheme for the deter- 
 mination of minerals, 279 
 
 Thurach's process for purify ing bis- 
 muth, 816 
 
 Tin, assay of, 537 
 
 argentiferous, 734 
 
 by wet way, 546 
 
 volumetrically, 550 
 
 - foil, 224 
 
 from antimony and arsenic, separa- 
 tion of, 564 
 
 in gun and bell metal, assay of, 549, 
 550 
 
 ore, 262, 285, 291 
 
 assay of, by fusion, 553 
 
 reduction with hydro- 
 gen, 552 
 
 ores, 287 
 
 containing arsenic, sulphur, and 
 
 tungsten, 543 
 
 oxide, 537 
 
 VOL 
 
 Tin oxide, assay of, 539 
 
 mixed with silica of, 542 
 
 slags, assay of, 553 
 Tinstone, assay of, 539 
 Tin-white cobalt, 269 
 Titanic acid in iron ores, 358 
 
 - iron, 267, 285 
 Titaniferous iron ore, 308 
 Titanium in iron, 430, 432 
 
 ores, 317 
 
 Titration of iron with potassium bi- 
 chromate, 321 
 
 sodium hyposulphite, 330,, 
 
 silver with ammonium sulpho- 
 cyanide, 711 
 
 Tongs, 67, 68 
 Topaz, 253, 292, 293 
 
 false, 250 
 
 - white, 876 
 
 yellow, 880 
 
 red, 887 
 
 blue 888 
 
 Tosh's process for the estimation of 
 
 graphite in iron and steel, 370 
 Touch-needles, 757 
 Touchstone, 757 
 Tourmaline, 252, 292 
 
 yellow, 880 
 
 brown. 886 
 
 - red, 887 
 
 blue, 889 
 
 violet, 890 
 
 green, 892 
 Turner's flux, 222 
 Turmeric paper, 213 
 
 Turner's process for estimating the 
 hardness of iron and steel, 433 
 
 estimation of silicon in iron 
 
 and steel, 413 
 
 the detection of boracic acid, 
 
 295 
 
 Turquoise, 889 
 
 Type metal, assay of, 565 
 
 TTNIVERSAL furnace, 65 
 U gas furnace, 84 
 Ure's calorimeter, 158 
 method, for assay of fuel, 157 
 
 TTANADATE of copper, 434 
 V Vanning, washing, or dressing, 
 23 
 
 Varvacite, 833 
 
 Vegetation of silver, 620 
 
 Vermeil garnet, 884 
 
 Vitreous copper, 273 
 
 282, 283 
 
 Volatile products of carbonisation of 
 fuels, 60 
 
 Volhard, titration of silver with am- 
 monium sulphocyanide, 711 
 
Ixii 
 
 INDEX. 
 
 VOL 
 
 Volumetric analysis, 298 
 assay of bismuth, 817 
 
 copper, 476 
 
 chromium, 829, 827 
 
 iron, 321 
 
 mercury, 595 
 
 silver, 634, 647, 674 
 
 - tin, 550 
 
 zinc, 575 
 
 Von Hubert's colorimetric assay of 
 copper, 472 
 
 WAEKEN'S process for assay of 
 cobalt speiss, 843 
 Washing, dressing, or vanning, 23 
 Water, distilled, 52 
 
 in fuel, estimation of, 152 
 iron ores, 351 
 
 minerals, 296 
 Waterproof cement, 116 
 Wax, yellow, 114 
 Wedgwood's pyrometer, 144 
 Weighing, 26, 36 
 
 moist precipitates, 39 
 
 Weight of minute silver globules, 723 
 
 spheres of gold, 753 
 
 Weights, 34 
 
 assay for silver and gold, 35 
 Wet assay of iron, 321 
 
 process for assay of silver, 634, 647, 
 674 
 
 _ l Z inc, 571, 572, 573 
 
 White flux, 197 
 
 lead, 176 
 
 and oil lute, 114 
 
 nickel, 278, 284 
 
 Wiborgh's process for estimating sul- 
 
 Shur in iron, 393 
 lis's lute for crucibles, 119, 124 
 Wilson's pyrometer, 147 
 Wind furnace, 57 
 
 Winkler's process for separating tin 
 from antimony and arsenic, 564 
 
 ZIR 
 
 Wohler.'s method of decomposing 
 
 osmiridium, 803 
 Wolfram, 278, 287 
 Wolfsbergite, 434 
 Wood charcoal, 171 
 Wood tin, 537 
 Wright's calorimeter, 159 
 process for assay of sulphur, 861 
 
 TTELLOW flame, 295 
 JL ochre, 267 
 wax, 114 
 
 7EOLITES, 255, 289 
 /J Zinc, 213 
 
 alloys, assay of, 574 
 
 argentiferous assay of, 735 
 
 assay of, 567 
 
 - by the wet process, 571, 572. 
 
 573 
 
 carbonate, 270 
 
 cobalt, and nickel, assay of, 850 
 
 coloured flame of, 296 
 
 copper and nickel alloys, assay of, 
 853 
 
 distillation of, 84, 568 
 
 from copper, separation of, 584 
 
 in iron ores,. 344 
 
 ores, 567 
 
 oxy chloride of cement, 120 
 
 reduction of iron salts by, 327 
 
 silicate, 270, 289 
 assay of, 572 
 
 sulphate, 288 
 
 sulphide, 187, 285 
 assay of, 573 
 
 volumetric assay of, 575 
 Zircon, 254, 292, 293 
 
 white, 875 
 
 yellow, 879 
 
 brown, 884 
 
 fpE 
 
 (fmVEBSr 
 
 '^ 
 
 PRINTED BY 
 
 SPOTTISWOODB AND CO., NEW-STREET SQUARE 
 LONDON 
 
 
UNIVERSITY OF CALIFORNIA LIBRARY 
 BERKELEY 
 
 Return to desk from which borrowed. 
 This book is DUE on the last date stamped below. 
 
 MAY 20 1948 
 
 LD 21-100m-9,'47(A5702sl6)476 
 
YC 68390 
 
 
 UNIVERSITY OF CALIFORNIA LIBRARY