IMAGE EVALUATION TEST TARGET (MT-3) 1.0 I.I '^i5 im 112.5 ... 1112 iiiM tt; 2.2 12.0 1.8 1.25 1.4 '6 — M 6" ■ — ► V2 <P^ % 4 & v^ e* ".r* yJ* O >■' >>. // y /^ Photographic Sciences Corpordtion ,%^' :^N^ <v \ \ 33 WEST MAIN STREET WEBSTER, NY 14580 (716) 872-4503 % .V o^ te' Mp< ^ CIHM/ICMH Microfiche Series. CIHM/ICMH Collection de microfiches. Canadian Institute for Historical Microreproductions Institut Canadian de micrcreproductions historiques 1980 Technical and Bibliographic Notes/Notes techniques et bibliographiques The Institute has attempted to obtain the best original copy available for filming. Features of this copy which may be bibliographically unique, which may alter any of the images in the reproduction, or which may significantly change the usual method of filming, are checked below. L'Institut a microfilm^ le meilleur exemplaire qu'il lui a 6X6 possible de se procurer. Les details de cet exemplaire qui sont peut-§tre uniques du point de vue bibliographique, qui peuvent modifier une image reproduite, ou qui peuvent exiger une modification dans la mdthode normale de filmage sont indiquds ci-dessous. r~7J Coloured covers/ I *l Couverture de couleur D D D D D D D Covers damaged/ Couverture endommagde Covers restored and/or laminated/ Couverture restaur6e et/ou pellicul^e Cover title missing/ Le titre de ouverture manque Coloured maps/ Cartes g^ographiques en couleur Coloured ink (i.e. other than blue or black)/ Encre de couleur (i.e. autre que bleue ou noire) Coloured plates and/or illustrations/ Planches et/ou illustrations en couleur Bound with other material/ Reli6 avec d'autres documents y / Tight binding may cause shadows or distortion __l along interior margin/ La reliure seir6e peut causer de I'ombre ou de la distortion le long de la marge intdrieure n Blank leaves added during restoration may appear within the text. Whenever possible, these have been omitted from filming/ II se peut que certaines pages blanches ajout^es lors d'une restauration apparaissent dans le texte, mais, lorsque cela 6tait possible, ces pages n'ont pas 6t6 filmdes. □ Coloured pages/ Pages de couleur □ Pages damaged/ Pages endommag6es □ Pages restored and/or laminated/ Pages restaurdes et/ou pellicul6es I — I Pages discoloured, stained or foxed/ Pages ddcolordes, taciietdes ou piqu^es Pages detached/ Pages d^tach^es Showthrough/ Transparence □ Quality of print varies/ Quality indgale de I'impression □ Includes supplementary material/ Comprend du materiel supplementaire D D Only edition available/ Seule Edition disponible Pages wholly or partially obscured by errata slips, tissues, etc., have been refilmed to ensure the best possible image/ Les pages totalement ou partiellement ubscurcies par un feuillet d'errata, une pelure, etc., ont §t6 filmdes d nouveau de faqon d obtenir la meilleure im>)ge possible. D Additional comments:/ Commentaires suppl6mentaires: ; ^ This item is filmed at the reduction ratio checked below/ Ce document est film6 au taux de r6duction indiqu6 ci-dessous. 10X 14X 18X 22X 26X SOX y taMi^^M 17X 16X 20X 24X 28X 32X The copy filmed here has been reproduced thanks to the generosity of: Library, Geological Survey of Canada L'exemplaire filmd fut reproduit grSce d la g6n6rosit6 de: Bibliotheque, Commission G^otogique du Canada The images appearing here are the best quality possible considering the condition and legibility of the original copy and in keeping with the filming contract specifications. Les images suivantes ont 6t6 reproduites avec le plus grand soin, compte tenu de la condition et de la nettetd de I'exemplrire film^, et en conformity avec les conditions du contrat de filmage. Original copies in printed paper covers are filmed beginning with the front cover and ending on the last page with a printed or illustrated impres- sion, or the back cover when appropriate. All other original copies are filmed beginning on the first page with a printed or illustrated impres- sion, and ending on the last page with a printed or illustrated impression. The last recorded frame on each microfiche shall contain the symbol — ♦- (meaning "CON- TINUED"), or the symbol V (meaning "END"), whichever applies. Maps, plates, charts, etc., may be filmed at different reduction ratios. Those too large to be entirely included in one exposure are filmed beginning in the upper left hand corner, left to right and top to bottom, as many frames as required. The following diagrams illustrate the method: Les exemplaires originaux dont la couverture en papier est imprim6e sont film6s en commengant par le premier plat et en terminant soit par la dernidre page qui comporte une empreinte d'impression ou d'illustration, soit par le second plat, selon le cas. Tous les autres exemplaires originaux sont film6s en commenqant par la premidre page qui comporte une empreinte d'impression ou d'illustration et en terminant par la dernidre page qui comporte une telle empreinte. Un des symboles suivants apparaitra sur la dernidre image de chaque microfiche, selon le cas: le symbole — »■ signifie "A SUIVRE", le symbole V signifie "FIN". Les cartes, planches, tableaux, etc., peuvent dtre film6s d des taux de reduction diff^rents. Lorsque le document est trop grand pour dtre reproduit en un seul cliche, il est filmd d partir de Tangle sup6rieur gauche, de gauche d droite, et de haut en bas, en prenant le nombre d'images ndcessaire. Les diagrammes suivants illustrent la m^thode. 32X 1 2 3 4 5 6 GEOLOGICAL SURVEY OF CANADA. ALFRED R. C. SELWYN, Director. NOTES ON IRON AND IRON ORES; BT . T. STERRY HUNT, LL.D., F.R.S., OHXUIBT AKD NINERALOOIBT TO THS 8DRVKT, J^XTRACTED FROM A REPORT ADDUK88£D TO SIR AVILLTAM E. I^OGAN, F.R.S., LATE DIBIOTOR OF THB BURVBT. From the Reports 0/ the Geological Survey of the Dominion of Canwlafor 1867-09. DAWSON BROS. : MONTREAL. n. WEHTERMANN k CO.: NEW YORK. SAMPSON, LOW, SON A MARSTON : LONDON. F. A. BROCKHAUS: LKIPSIO. BALLlbRK; PARIS. Ib70. V I n mm GEOLOGICAL SURVEY OF CANADA ALFRED R. C. SELWYN, Director. NOTES ON IRON AND IRON OR BT T. STERRY HUNT, LL.D., F.R.S., CHEMIBT AJiD MINERALOGIST TO THE 8UBVEV, EXTRACTED FROM A REPORT ADDRESSED TO SIR WILLIAM E. LOG- AN, F.R.S.. LATE DIRECTOn OK THE 8CBVBV. From the R'tports of the Geological Surveij of the Dominion of Canada for 1867-6!). DAWSON BROS. : MONTRRAL. B. VVESTER.MANW k CO.: NEW YORK. SAMPSON, LOW, HON k MARSTON ; LONDON. F. A. BROCKEIAUS: LEIl'SIC. BALLlfcRE; PARIS. 1870. t * « • • • • • ••• • • ••• . • * • • - • • • • • • . Ml TABLE OF CONTENTS. Iron ores op Swedes, Norway and Canada compared I Mining ores and making clmrcoal 2 Weight of chBrcoal ; its cost 3 Mining laws In Sweden 4 Sulphur in iron ores ; roasting 5 Pliosphorus in ores ; manganese ; titanium 6 Ores of Dannemora and Taberg ^ Working tiianiferous ores 8-9 Analyses of some iron ores 9 Hull ores ; Hull blast-furnace : cost of smelting 10-11 American blast-furnaces ; St. Maurice 12 Ore of South Crosby 13 Ores of Xorih Crosby and Helraont 14 Ores of Madoc and MacNab 15 Ores of Gros Cap and Hay of Seven Islands 16 Iron sands j their origin ; gold alluvions 17 Great lakes, St. Lawrence, Atlantic coast 18 Early workings ; Elliot in Connecticut; Naplea 19 Moisie sand-ore, occurrence, extraction, specific gravity 20-21 Bersimis, Seven Islands, Mingan, etc 22 Magnetic separation ; analyses 22-25 Making iron and steel by direct methods 2G Catalan, Corsican, Osmund iind Genoese forges 27-28 German or American bloomary forge 29 Advantages and use of bloomarics in the United States 30-31 l'r('i)aration of ores ; Palmer ore-bed 32 Construction of bloomary hearths 33' Mode of working ; New Russia and East Middlebury forges 34-35 Moisie forges ; comparative results, nature of sand-ore 3()-37 Impurities ; fluxing, loss of iron ; analyses of slags 38-39 Quality of Moisie iron ; cist of making blooms 40-41 Direct and indirect methods compared 41-42 Purifying oies ; Larue's magnetic separator 42-43 Direct processes ; Chenot's ; its cost ; Clay's and Renton's 44-45 Harvey's ; Guvlt's plan ; carbonizing iron by gases 46 Plans for working sands; Stenson, Moreau, Hodges, Wliclpley and Storer. 4 7 Treatment in crucibles ; Ponsard, .lohnson, Larue 48 Cast steel ; Heath's patents ; Ucliatius, Obuchow 49 Martin, Hessemer, Heaton 50 Sicmens's direct process ; Leckie's process 51 Siemens's regenerai i ve gas-furnace 52 Burning wet fuel ; the Lundin furnace 53 Boctius's gas-furnace 54 The Ellersiiausen process for Mali-karle Iron. . 54 Malleable castings ; old Welsh process, Tunaer's, Ellershausen's discovery 55 Theory of Eller»liau3cu's process 56 Dr. Wuth's analyses ; conclusions 57 Suggestions for practice ; clioic of ores 58 Mode of working; advantages of the process 59 Use of granulated cast iron ; Hewitt's results 60 T T NOTES ON IRON AND IRON ORES, BY T. STERRY HUNT, LL.D., F.R.S. The following pages are extracted from the Reports of the Geological Survey of Canada for 1867-69, and form a part of the author's report,* dated Nov. 1, 1869, and addressed to Sir William E. Logan, F.R.S. , late director of the Geological Survey of Canada. They have been written especially for the design of aff ling information with regard to some of the iron ores of the provinces of Ontario and Quebec, and the best modes of working them. In a subsequent report, Ht is proposed to continue the subject, and to add details respecting the ores of New Brunswick and Nova Scotia. The reader will find, prefixed, a table of contents, which will serve to shew the points herein discussed. The iron ores of the Laurentian system are, for the greater part, of the or.>8 of Norway '' ' o 1 . 5„|i .Sweden. magnetic species, and are similar in geological relations and in mineralo- gical characters to the ores which occur in the same system in northern New York, and in the Highlands of southern New York and New Jersey, where they have long been mined to a great extent. Similar ores, more- over, abound in Norway and Sweden, where they occur in rocks of the same age, and furnish great quantities of very pure iron, which is famous throughout the markets of the world. Having had opportunities at the Exposition at Paris, in 1867, to learn many facts about the iron-industry of these countries, I have thought it would be well to embody some of them in the present report, as likely to prove valuable to the mining interests of the Dominion. A large portion of both Norway and Sweden is occupied by old gneisses of the Laurentian system, which also comprise the greater part of the provinces of Ontario and Quebec. This geological resemblance, with somewhat similar conditions of soil and climate, gives to any facts relating to the mineralogy and metallurgy of those northern regions, a special interest to the people of Canada. In the year 1865, according to oflScial data, there were extracted in walking, Sweden 492,474 tons of iron ore, employing 5,062 workmen. The mines or openings from which this amount of ore was raised, are stated to be 524 in number, and some of them are evidently worked on a very small scale. The workings are ordinarily by open cuttings upon the beds or masses of ore, which are described aa being very generally in a nearly vertical atti- • The numbers at the heads of the pages, in the outer corners, commencing with 245, correspond with the pages of the volume from which these notes are extracted, and are referred to in the text ; while the numbers in the inner corners are those of the prefixed table of contents. Charcoal making. 246 GEOLOGICAL SURVEY OF CANADA. [2 tude, and in solid crystalline rock, recjuiring but little support by tim- t boring. The mineral is mined with powder, although nitro-glycerine ha:^ been tried to some extent. The pay of the workmen ranges from thirty to fifty cents per day, and the cost of the ore, when raised, is said to vary from ' one to two dollars for the ton of 1000 kilogrammes (2205 pounds avoirdu- pois). With the exception of a small quantity carried into Finland, the whole of this ore is smelted in th ountry. The production of iron ores in Norway is much less than Sweden ; about 22,000 tons are raiseil annually, of which 2500 tons are exported, the remainder being smelted in blast-furnaces with charcoal. At one of the most important of these, that of Laurvig, where a remarkably fine iron is made for the American market, the cost of the ore at the furnace is stated at §1.80 the ton. ^ In Sweden, and in Norway, charcoal is the only fuel employed for the I reduction of the iron ores, except in some rare instances, where a mixture " of charcoal and dry wocd has been used in the blast furnace. Careful trials, however, appear to show that this admixture offers no advantages over the use of charcoal alone. About one-third of the surface of Sweden is covered with forests, which constitute an important source of wealth to the country, and of late years have been the object of care and attention, with a view to a due economy of fuel and lumber. The trees of the Swedish forests, with the exception of the southern peninsula, where oak and beech are met with, are chiefly of coniferous or sof^woodcd species, and the pine of the country (Pinus sylvestn%) is the one principally used for metallurgical purposes, the timber being sawn or hewn for lumber, while the branches are employed for the manufacture of charcoal. The wood is cut in the months of March and April, before the I'ising of the sap, and is divided into lengths of about eight feet, which are allowed to dry during the summer months. The charcoal-burning takes place in October and November, and is generally carried on in circular piles about twelve feet ■high and from twenty to thirty feet or more in diameter. The burning of a pile lasts from two to three weeks from the time of kindling. Expe- rience has shown, in Sweden, that the economy is much greater when the wood is laid upon its side in the piles than when placed on end. In the latter case the yield of charcoal is from 60 to 62 per cent, of the volume of the wood, while in the former it is not less than 70 per cent. According to a Report to the Swedish Minister of Agriculture, Commerce and Pub- lic Works, published in 1866, the average cost of labcr for a pile yield- ing from twelve to thirteen tons of charcoal, is 8-4 francs, which is e(iual to about f 1.30 for the ton of 1000 kilogrammes. This price includes the cutting and drawing of the wood. The cubic meter or stere of 35'317 cubic feet of pine charcoal in Sweden flveighs from 142 to 145 kilogrammes, so that the ton of 1000 kilogrammes -1 ar 81 he us Clll thj ac| Ok 1 [2 le support by tim- 1 aitro-glycerine has nges from thirty to s said to vary from 5 pounds avoirdu- into Finland, the iction of iron ores D tons are raised ler being smelceil iportant of these, for the American .80 the ton. employed for the where a mixture arnace. Careful rs no advantages irface of Sweden rce of wealth to re and attention, !3 of the Swedish where oak and led species, and icipally used for 1- lumber, while 1- The wood is ' the sap, and is i to dry during in October and )Out twelve feet The burning of idling. Expe- eater when the 1 end. In the f the volume of t. According lerce and Pub- i' a pile yield- liich is ecjual to !e includes the SJ REPORT OF DR. T. STERRY HUNT. 247 I !oal in Sweden ) kilogrammes I (2205 pounds) would measure very nearly 7 steres, or 247 cubic feet, and the weight of the cubic foot of charcoal would be a little over 4 kilo- charcoal, grammes, or 8.8 pounds, nearly. According to figures given by Grill, however, (Percy, 3Ietallurgy of Iron, page 596) a ton of the charcoal used in the Lancashire hearths, in Sweden, measures not loss than 297 cubic feet. In the American iron-regions charcoal is bought and sold by the bushel, which is an arbitrary measure of about five pecks, equal, according to Overman, to 2000 cubic inches, and according to Osborn to 2675 cubic inches, (the United-States standard, or Winchester bushel, measuring 2150.42 cubic inches.) Taking the latter figure, we find that the American charcoal-bushel of Swedish pine-charcoal would weigh a little over 13.5 pounds avoirdupois. The experiments of Francois, in the Pyrennees, give for the weight of the cubic meter of charcoal of beech and oak, from 218 to 235 kilogrammes, that of alder being 141, and that of pine and spruce from 152 to 173. lie deduces as the mean for hard-wood charcoal 227, and for soft-wood, 170 kilogrammes, corresponding respectively to 21.9 and 16.4 pounds avoir- dupois for the charcoal-bushel as above. (Jules Francois, i)es Min- erais de Fer, etc., page 177.) The elaborate studies of Mr. Marcus Bull on the charcoal from North American woods, give the following as the weights, in pounds, of a bushel of dry charcoal from these kinds, among others : red cedar 12.52, white pine 15.42, yellow pine 17.52, white birch 19.15, and several varieties of maple and oak from 21 to 23 pounds.* This last is confirmed by the observation of Mr. Kennedy, at the Hull Iron- works, who informed me that a bushel of mixed beech and maple, such as there used, weighed from 22 to 23 pounds. The cubic meter is equal to about 22.8 charcoal-bushels of 2675 cubic inches, and the price of the cubic meter of charcoal, which reaches at some furnaces, $1.30, is on an average, in Sweden, 85 cents, or abort tostofcoai. four cents the bushel. At the iron furnace of Laurvig, in Norway, the cost of good charcoal is said to from to 60 to 70 cents the cubic meter. In a few localities in Sweden, where water-courses afford facilities for floating the wood to the furnaces, the charring is effected in ovens of a peculiar construction, furnished with an arrangement for condensing the acid and tarry products given off during the process. The plan of one of these furnaces, shown at Paris, in 1867, was similar to that figured by Dr. Percy, on page 125 of his first volume on Metallurgy, in which will be found discussed in great detail, the whole subject of charcoal-burning, on pages 107-142. • Tbese results were published in the Transactions of the American Philos. Society, for 1826, new series, pp. 1-GO, and are reproduced in the American edition of Knapps Technology, i, 24. 248 GEOLOGICAL SURVEY OF CANADA. [4 Law concerniii;; mines, Although the Swedish ores vary considerably in their richness, it may be calculated that, in general, about two tons of ore are required for one ton of cast iron, to produce which are consumed on an average about a ton of charcoal. It is evident ther3fore that, for the same cost of pro- carryingftiei. duction, the fucl Can be transported much farther than the ore. Charcoal is often carried from localities where wood is abundant, to blast-furnaces in the vicinity of mines, a distance of twenty or thirty leagues. This is done in part by water or by rail, but for the transport of the ores from regions not easily accessible at other times, sledges are much used in the winter, which becomes the most favorable season for getting both the charcoal and the ores to the furnaces, which are generally as near as possible to the mines. In some cases the ores are carried for distances of ten or more leagues ; but this is generally when there is a back-freight of iron or other materials. The wages of a carter, with his horse, vary from $0.80 to #1.40 per day, and the cost of transporting the ore is from 6t\ to 9y% cents the ton for the English mile. The law with regard to mines in JSweden is as follows : The discoverer becomes the owner of one-half, while the other half remains the property of the owner of the land, who can work it by sharing the cost with the dis- coverer, or dispose of his share in the mine. A permission to work a new mine must be given by the magistrate ; and if left unworked during a cer- tain number of years, without obtaining a special authorization from the magistrate to do so, or without performing annually an amount of labor, stipulated as necessary to retain possession of the mine, the permission lapses, and the mine can be taken up again by another party on the same terms as a newly discovered one. Many of these mines are worked on a small scale, by little proprietors, who sell their ore, or in other cases join their forces and construct, between them, a blast-furnace at a cost of from 112,000 to $14,000. Much of the iron manufactured in Sweden has, from the earliest period, been in the hands of peasants and small proprietors. The manufacture of cast iron in Sweden goes back aboiit 200 years ; previous to that time wrought iron was made from the ore by a diroct method. Those regions where ore and fuel furnished conditions favorable to mining industry, were formerly consti- tuted into districts, which were invested by the state with certain privi- leges, and subjected to certain restrictions, one of which was to export beyond their limits all the cast iron manufactured within their respective districts. All of these restrictions aie now, however, abolished. Biatt-fiirnacu". The total number of blast-furnaces in Sweden is about 300, of which 219 were in blast in 1866, and instead of being grouped together, as in some other countries, they are, with few exceptions, isolated ; a single furnace being erected in some spot where a wator-power and facilities of a bu| 491 full or( tlif| 3ull pre add [4 leir richness, it may ire required for one an average about a e same cost of pro- the ore. Charcoal to biast-furnaces in gues. This is done J ores from regions ised in the winter, th the charcoal and as possible to the !es of ten or more ;ht of iron or other )ma0.80to#1.40 r to 9^V cents the ' : The discoverer ns the property of cost with the dis- ion to work a new fed during a cer- rization from the amount of labor, N the permission arty on the same ittle proprietors, iistruct, between 000. Much of 'iod, been in the 3 of oast iron in Tought iron was ere ore and fuel formerly consti- 1 certain privi- was to export heir respective ihed. 300, of which together, as in ited; a single itnd facilities 5] REPORT OF DR. T. 3TERRY HUNT. 249 v)f transportation are met with in proximity to forests sufficient to afford a supply of charcoal, the deposits of ore being pretty widely distri- buted. The amount of ore raised in 1865 has been already stated at 492,474 tons, employing 5063 workmen. The production of the various furnaces in the same year was 226,676 tons of cast iron, employing 1 3683 workmen, whose wages ranged from 80.30 to $1.40 per day. Composition of ore*. The ores vary in richness from the nearly pure magnetic or specular ores, containing as much as 70 per cent of iron, to those yielding not more than 28 per cent. The Swedish ores and irons have been made the subject of very minute and extended chemical studies, with reference to the proper composition of the charges, the nature and quantity of fluxes to be ,; added, the various impurities in the ores, and the influence of all these I upon the quality of the iron. Foremost in importance are considered the influence of sulphur, phosphorus and manganese. Both sulphur and I phosphorus are regarded as especially detrimental to the iron destined for ' the forgo, or for the manufacture of steel, and from these impurities the Swedish ores are generally very free, when compared with the ores of England and France, a purity which they may be said to share in common with the Laurcntian ores of North America. The observations which have lieon made with regard to the Swedish ores, in this respect will, therefore, for the most part, be equally applicable to our own. The sulphur of the suiphur. Swedish ores is generally present in the form of pyrites or sulphuret of iron, and may be expelled by resisting at a red heat, which completely oxydizes this substance. If, however, carbonate of lime is present at the same ; time, a portion of sulphate of lime is formed, by which some of the sulphur is retained, and can only be removed by subsequent washing with water, in which the sulphate is slightly soluble. It does not appear whether the use of water is ever thus resorted to. The ingenious furnace of West- ciiciuation. inann, by which tlie waste gases from the blast-furnace are employed to effect the roasting and desulphurizing of the ore, is said to have been found thoroughly efficient in Sweden, and is now in use at Kingwood, in New Jersey, in connection with a blast-farnace, by Messrs. Cooper, Hewitt and Co. In some cases the roasting of the ores in Sweden is two or throe times repeated. The heat is so greai that they are more or less softened, and show a commencement of fusion. The magnetic ores, after this process, appear to be more readily rcduood than before, though the roasting seems, from the result of analyses at Fahlun, to have but little affected tlio state of oxydation of the iron. The favorable effect is probably duo, in part, to the Assuring of the ore by the heat. The presence of even small portions of sulphur in wrought iron renders it, as is well known, brittle when hot, or red-short, as it is termed. For certain purposes, however, the presence of sulphur in cast iron is not objectionable. Thus, for casting 260 GEOLOGICAL SURVEY OF CANADA. re ( cannon, according to Rinman, a very strong metal is obtained by adding to the charge a small amount of suliihuret of iron, and in general for this purpose a charge is preferred free from phosphorus, but somewhat sulphu- rous. The sulphur causes a larger proportion of carbon to remain in a combined state ; a very tenacious mottled cast iron is obtained, holding about 0.09 per cent of sulphur, and the quantity may even rise to 0.30 or 0.50 per cent. The use of sulphurous ores, according to Rinman, like that of manganesian ores, enables us to obtain ^Yhite iron when the furnace is running at its ordinary rate, and without any overcharge of ore. rhosphonis. Phosphorus, in like manner, though it renders wrought iron cold-short, gives to it a hardness which renders it peculiarly valuable for some purposes, as for boiler-plates, roofing-sheets, spades, shovels and hoes, and other utensils which are exposed to severe wear. In the metal for these, at least 0.1 per cent of phosphorus, and in that for fine castings as much as 0.5 per cent, is considered advantageous, as contributing in the latter case to give greater fusibility and fluidity to the melted metal. But for the manufacture of steel, phosphorus seems to exert a highly prejudicial influence, and it appears from carefully-made analyses of Swedish irons, that their value in the Sheflfield market, where their relative fitness for the manufacture of steel has been determined by experience, is, as shewn by Rinman, directly in proportion to their freedom from phosphorus. The amount of phosphorus in the ores of Dannemora, Bispberg, and some other of the Swedish mines does not exceed 0.005 per cent., while in some others, as Gellivara and Graengesberg, it rises to 1.3 and even 2.0 per cent. Some of these ores, like similar ores in northern New York, contain imbedded grains of phosphate of lime or apatite. It is, however, to be remarked that the whole of the phosphorus in the charge does not pass into the ores, and moreover, that the proportion of this element varies in diflfcrent parts of the deposit, so that liy a judicious admixture of the phosphurctted with purer ores, the resulting cast iron will not contain more than 0. 15 per cent, of jihosphorus, which does not render it unfit for ordinary uses. Mangnnesf. Manganese is also conceived to exert an important influence, in more ways than one, upon the (juality of iron. The Swedish ores not unfre- (lucntly contain a jiortion of this element, and when absent from any ore it is sought to be supplied by mixtures containing manganese. While the greater part of it passes intu the slags, a certain portion remains in the cast iron, and to its presence it is customary to ascribe a jjeculiar fitness in the resulting malleable iron for the manufacture of steel. It is, however, remarked that manganese is often wanting, without any observed inferior- ity in the cast iron. The presence oi titanium, and its influence upon iron, is a subject which has of otl st a on U'O [6 ned by adding general for this lewhat sulphu- :o remain in a ained, holding n rise to 0.30 ' Rinman, like !n the furnace )f ore. ■on cold-short, ble for some ind hoes, and 3tal for these, ;3 as much as he latter case But for the 7 prejudicial wedish irons, itness for the as shewn bj rus. ispberg, and 3nt., while in even 2.0 per fork, contain wever, to be loes not pass ent varies in dure of the not contain r it unfit for ice, in more 3 not uufre- •om any ore While the 3 in the cast tncss in the i, however, ed inforior- t which has 7] REPORT OF DR. T. STERRY HUNT. 251 of late been very much debated. While claimed by Mr. Mushct, and some others, to exert a special and most beneficial influence on the quality of steel, this is denied by others. When ores containing titanium are smelted, Titanium, a small portion of this element, amounting in some cases to a little over one per cent., passes into the cast metal, and is said to increase its strength, besides giving it a peculiar mottled aspect. It seems, however, " doubtful whether any titanium remains in the bar iron or steel made from such pig iron, so that the improvement attributed to the use of titaniferous ore is probably due to some indirect action, rather than to the actual presence of titanium in the finished product. The evidence on this point is not suf- ficiently clear to allow of any positive conclusion being formed." To the above statement of Bauerman, I may add that I have failed to detect any titanium in bloom iron made by the direct method from an iron ore con- taining 16 per cent, of titanium, which will be described further on. Some remarks upon the composition and the results of analyses of the Swedish ores may not be without value, as serving for comparison with the iron ores of Canada. The iron, both of Sweden and of Norway, is made, with but few exceptions, from ores of the magnetic species. That of the famous Dannemora district, which supplies a great number of blast-fur- JJ.^'in*™"" naces, and produces an iron regarded as superior to all others for the manu facture of steel, occurs as an irregular interrupted belt, a mile and a half in length, which is imbedded in crystalline limestone, with a kind of petrosili- cious rock, and has been mined to a depth of more than 100 fathoms. The composition of difl'erent portions of the deposit presents considerable varia- tion. Average specimens from one of the most important masses, sent ta the Paris Exhibition in a roasted state, as prepared for the furnace, showed considerable admixtures of silica, lime and magnesia, with some alumina. The sum of the united protoxyd and pcro.\:yd of iron for these two ores, was respectively about 54 and 08 per cent., equalling 88-5 and 48-0 per cent of metallic iron. These two ores were almost destitute of sulphur and phosphorus, and had the advantage, when mixed, of yielding .1 fusible slag without the addition of any limestone for flux. Others of the Swedish ores are much richer in iron than these, while others, still, are very much poorer. Thus, at Taberg, an ore is mined, which tuIjitb ore. consists of magnetic iron disseminated through a serpentine, (sometimes described as a diorite), the magnetic oxyd constituting not more than one half of the mass. This ore, which contains at the same time, from t! to 10 per cent, of titanic acid, yields only about '2o or oO per cent of iron. It is melted with about one-fourth its weight of limestone as a flux, and gives a white mirror-like ca«t metal, which yields an iron much esteemed for wire-drawing. Fuel being cheap in the neighborhood, tliis ore is exten- sively mined and smelted. Bauerman states that attemjits were made to . 252 GEOLOGICAL SURVEY OF CANADA. [8 I 9] ■Working titauic oriM. treat this ore, previously dressed so as to yield 43 per cent of iron, but for this purpose it was necessary to bring it to such a finely divided condition, that it was judged better to smelt it in its natural state, the expense due to the increased consumption of fuel, being counter-*: alanced by greater faci- lity in treatment. Besides this of Taberg, other similar ores have long been smelted in Norway and in Finland. The ore from the Cristine mine at Krageroc, in southern Norway, is described as a brilliant black titanife- roiis magnetite, not very strongly attracted by the magnet, and intermixed with grains of quartz, and of greenish-black hornblende, with a little magne- tic pyrites. It contained no phosphorus, but gave by analysis 42.0 per cent, of metallic iron, besides 15.10 of titanic acid and 19.9 of silica, with a small amount of earthy bases. Inasmuch as many of our Canadian ores are more or less tltauiferous, the following notes with regard t > the smelting this and other titauiferous ores are of much interest. They are extracted from a communication by Mr. David Forbes, in the Chemical Neivs for Do raber 11, 18GS. " The experience of the Scandinavian iron-masters has shown that the only objection to the use of titaniferous ores is that they are found to be more refractory in the blast-furnace, in proportion as they contain a greater percentage of titanic acid ; and if much titanium is present they require a so much larger amount of charcoal to smelt them as not to render their employment profitable in a country where other ores free from titanium can be obtained at a reasonable rate. After considerable expe- rience in smelting the ore of Krageroe, which yielded a very good iron, it was found unprofitable to smelt it alone, for the above reason ; but its use was found beneficial when employed in about equal proportions with the other ores of the district, which were free from titanium." Mr. Forbes found, in his experience, that by employing a mixture of crushed quartz and limestone as a flux, when the proportion of titanium in the ore did not exceed eight per cent, or was reduced to this amount by admixture of ores free from titanic acid, no difficulty was experienced in working this ore cleanly and profitably. The iron produced was free from phosphorus, gave but a trace of sulphur, and only <)'05 of titanic acid, which was sup- posed to be mechanically present rather than chemically combined with the iron. Another very similar ore from Eger, which contained 38'89 per cent, of iron and T'lO of titanic acid, was found to contain too much sulphur and phosphorus to be fit for bar iron, but yielded a good foundry-pig metal, which gave by analysis 0'2C) of titanic acid. When smelted alone it •was refractory, and did not yield a liquid slag, but it was readily fused when mixed, as at Krageroe, with ores destitute of titanium. The experience of the irou-masters in New York, who have endeavored to smelt the titaniferous ores of Lake Champlain, generally in admixture wil sul eni ne| coil onl viil 111? inc reel se^l ort ma ad [8 ; of iron, but for vided condition, he expense due bj greater faci- ores Lave lonit e Cristinc mine t black titanife- and intermixed I a little raagno- lysis 42.0 per ' of silica, with our Canadian regard to the St. Thej are the Chemical hown that the e found to be they contain m is present hem as not to ares free from -lerable expe- good iron, it ; but its use ons with the ^Ir. Forbes . quartz and ore did not admixture of ing this phosphorus, ich was sup- ned with the 38'89 per n too much foundry-pig ted alone it fused when endeavored admixture w 9] REPORT OF DB. T. STERRY HDNT. 253 with other ores, has been very unfavorable, but an attention to the above suggestions might probably enable them to overcome the difficulties hitherto encountered. Besides the great bed of ore at Bay St. Paul, holding nearly half its weight of titanic acid, Canada has large deposits of ores containing more or less titanium, some of which will be described farther on. In the G-eology of Canada, page 501, I have shown that a massive Titanic ores, granular titaniferous ore from St. Francois, on the Chaudierc, in the pro- vince of Quebec, consists of a mixture of about two-thirds of nearly pure magnetic oxyd of iron, and one-third of a titanic iron or menaccanite hold- ing not less than 48 per cent of titanic acid. Tlie two are, however, readily separable by a magnet, and it is probable that by a magnetic separating machine it will be possible to make use of this and of similar ores for the preparation of iron in the direct way, to which tlie purified magnetic oxyd is well adapted. The iron sands, which contain a large admixture of titanic iron, will be noticed in their placr. In this connection I quote from Osborp's recently published volume on the Metallurgy of Iron and Steel, page 47 J, the following statements, which he gives as a communication from a Mr. Henderson, according to whom an ore from Norway, holding over 40 per cent of titanic acid, is now sue cessfully smelted at Norton, in England, by a process patented by Player of New York. The ore is said to be smelted in small furnaces, with a blast at lOOO'' temperature ; 2 tons of coal being required to 2i tons of the ore, with 15 cwt. of limestone, and about 10 cwt. of basalt. The pig-metal thus produced is stated to contain very little carbon, and to be very easily puddled, producing a malleable iron of great tensile strength. Such ores are necessarily poor in iron, as compared with magnetic ores, and even if they can be readily smelted by the above treatment, it remains to be seen whether their use offers any real advantage. ANALYSES OF SnMK IRON ORES. The bed of magnetic ore, which has long lieon known at Ilidl, is de.>- crilied in the (.rcolomj of Canada, page 074. TIic association of a portion of a red hematite with tlie magnetic ore, and of graphite with both, is des- cribed in the Report of the Survey for 1800, p. 210. Since then a large uuii, oniaiio. blast-furnace has been erected here, which for some time produced a superior (piality of pig-iron ; but the working has been since aliandoned, the economic results not being satisfactory. The two samples whose analyses are here given had been prepared for that purpose hy Mr. Kennedy, the director of the works, and selected so as to represent the average of the ore smelted. One of these, designated at the furnace as the red ore, was ci.ilnrcd by an admixture of hematite, while the other was known as the black ore. The red ore jiave as follows : — I 254 Hull, red ore. GEOLOGICAL SURVEY OF CANADA. [10 Perosjd of irou C6.20 » Protoxjdofiron 17.785 = metallic iron 58.78. Oxyd of manganese traces. Lime, as silicate 76 Magnesia, as silicate 45 Carbonate of lime 2.66 Silica 10.44 Graphite 7i Pliospliorus 015 Sulphur 280 99.295 11' in Call witf fl The black magnetic ore of Hull contains a considerable amount of silica, together with a portion of a hydrated silicate of iron and magnesia, which causes the ore to yield an olive-brown powder. When the magnetic portion is removed from the pulverized ore by a magnet, there remains a considerable proportion of dull olive-colored earthy matter, which gives a pale brown streak, and is readily attached by hydrochloric acid, with sepa- ration of flocculent silica. In the following analysis of an average sample of the ore the whole was treated together, and all of the iron is represented as magnetic oxyd. Neither of the ores from Hull yielded any titanic acid. and the black ore contained neitlier lime nor manganese. It gave .Magnetic oxj'd of iron 73.90 = metallic iron 53.20. .Magnesia • 1.88 Ahiminn .61 Silica 20.27 Water 3.27 Pliosphorus 027 Sulpiiur 085 100.042 The height of the Hull blast-furnace is 38 feet, its diameter at the boshes being 10 /j feet, and at the throat 4 -^^ feet ; the twyers are six in rurnnce-chflrpo number. The charge at the time of my visit, in August 1868, consisted of 19 bushels of hard-wood charcoal, 460 pounds of the above ores, previously calcined, and mi.xcd in equal proportions, and 110 pounds of flux consisting of white crystalline limestone Go, clay 27, and silicious sand 18 pounds. The furnace was then yielding gray pig-iron, at the rate of 56 per cent for the ore, while the consumption of charcoal for the ton of metal, was 170 bushels. This was made from beech and maple, and as I was informed by Mr. Kennedy, Aveighed from 22 to 23 pounds to the bushel, beuig at the rate of 34 or 35 cwt. of charcoal to the ton of iron. The furnace was for a time in blast in 1867, and for a longer period, [10 111] metallic iron 58.78. REPORT OF DR. T. STERRY HUNT. 255 in 1868. By the kindness of Mr. Phillip S. Ross, the secretary to the Canada Iron-Mining and Manufacturing Company, I have been furnished with a statement of the -working results during that season. The furnace was in blast from April 27 to October 5, 1868, or 163 days, during which time there were coiisumed as follows : — Working at Hull. imount of silica, magnesia, which the magnetic bere remains a which gives a icid, with sepa- verage sample is represented ny titanic acid, It gave lie iron 53.20. meter at tlie ers are six in , consisted of s, previously ux consistiiii: pounds. The cent for the 170 bushels. lied by Mr. at tlio rate Igor ]ioriod. Hull ore 1835^;'^ tons. ^ Arnprior (McNab) ore eOjf"^ tons- J 1896 tons. Scrap iron T^^j " Limestone (clay and sand not estimated) 211 " Charcoal, soft wood at 4jc 133.573 bushels. % " hardwoodatSc 95.947 " \ 242,782 bushels. " mixed wood at 5Jc 13.262 " ) Wood at $1.25 25J corda. Peat, 80 tons, yielding of coke 21^?, tons. Pig iron produced 1040-/u- " The cost of the iron thus produced was as follows, per ton : — For ore, fuel, and wages of men S22.C0 Salaries and general expenses 3.10 Cost of a ton of pig iron at Hull S26.50 If we deduct from the total amount of metal produced, the scrap iron added, we obtain, as the average results during the season of 18G8, the following figures : — Daily production of pig iron 6^ tons. Yield of ore per ton 54.5 per cent. Charcoal consumed per ton of iron (at 5/|f(, cents.) 235 bushels. Peat-coke " " " 47 pounds. If we leave entirely out of the account the amount of peat-coke, and take the average weight of the charcoal at 18 pounds to the bushel, we shall have a consumption of 373 cwt. of charcoal to the tou of iron, while, with hard-wood charcoal, there were consumed, as above, from 34 to 35 cwt. In Sweden, according to Baucrman, the average consumption of charcoal, for the whole country, is from 16 to 17 cwt., for the ton of white or mottled pig iron, and about one-third more, or from 21 to 22 cwt., for the ton of gray metal suitable for foundry purposes or for Bessemer steel. At Langshytta, the consumption is as low as 13i to 14 cwt,, for the produc- tion of white or mottled iron, while the very poor ores of Taberg, already referred to (page 251), where the charge contains only 20 per cent of ron, require as much as 50 or 60 cwt. of charcoal per ton. 256 GEOLOGICAL SURVEY OF CANADA. American furnaces. Hull iron. [12 At the Greenwood furnace, near Marquette, on Lake Superior, is a charcoal furnace in which the unroasted ores of the region are smelted with a little crystalline limestone for flux, and yield 55 per cent of iron. To produce a ton of gray pig iron are consumed 140 bushels of charcoal, chiefly of maple, weighing from 16 to 20 pounds each, or about 23 cwt. of charcoal. At the Wyandotte works, near Detroit, where the red slaty hematite of Lake Superior is smelted, and yields on an average 65 percent of iron, there are consumed 140 bushels of soft-wood charcoal, weighing 14 pounds to the bushel, or 17^ cwt. to the ton of iron. (Bauerman, Metallurgy of Iron, 1^. 206). The recent returns from American blast- furnaces, published by Prof. Egleston, of the School of Mines, New- York, show that while many American charcoal-furnaces are still working in a very wasteful manner, the consumption of charcoal in some in New York and Michigan, is as low as lOU and 105 bushels. At the large blast- furnaces of Port Henry, on Lake Champlain, where magnetic ores similar to that of Hull are smelted with anthracite coal, the average consumption is from 1.10 to 1.14 tons, equal to 22 or 23 cwt. of anthracite to the ton of pig iron. With these facts before us, it is clear that the rich ores of Hull, with proper management, should be smelted with 22 or 23 cwt. of charcoal, instead of from 35 to 38 cwt., the quantity actually consumed. This alone is sufficient to explain the failure to produce iron profitably at Hull, where the supply of rich ore is abundant, and the quality of the iron made was excellent. It is evident from the analyses of the ores above given that the addition of sand and clay to the charge was unnecessary, and that limestone alone, in proper proportion, would have been sufficient for the purposes of a flux. A series of samples of pig iron made at the Hull furnace, was taken by me for analysis, but the results not being yet complete, are reserved for a future report. It may be stated however that a sample of the white iron made Avith a mixture of peat-coke and charcoal, contained 0.085 of phos- phorus and 0.28 of sulphur. This amount of sulphur may be due to the considerable proportion which, in the form of sulphate of lime, I have found in the ashes of some Canadian peats. St. Maurice. — In the well known blast-furnaces of Messrs. McDougall at St. Maurice, near Three Rivers, in the province of Quebec, where the bo^' ores of the region are smelted with a hot blast, the charge consists of 500 pounds of ore, with 25 pounds of limestone, and 16 bushels of mi.xed charcoal. The results for the month of December, 1868, showed a consumption of 26,272 bushels of charcoal and 372 tons of ore, with a yield of 163i tons of iron, of which about eleven-twelths were soft grey pig. This gives a production, for the ore, of 43 per cent of iron, with a L3] :onsl \nal] he ess fysisl letl [12 13] ake Superior, is a ■egiou are smelted 5 per cent of iron, ushels of charcoal, or about 23 cwt. of lere the red slaty /erage 05 per cent :liarcoal, weighing ■on. (Bauerman, a American blast- ^lines. New- York, still working in a •mo in New York Lt the large blast- )tic ores similar to ge consumption is cite to the ton of )res of Hull, with cwt. of charcoal, C'lisumed. This rofitably at Hull, ality of the iron that the addition limestone alone, irposes of a flux, e, was taken by e reserved for a f the white iron I0.085ofphos- y be due to the 'f lime, I have 5. McDougall at 5CC, where the hargo consists 10 bushels of 8(38, showed a of ore, with a ivore soft grey r iron, with a REPORT OF DR. T. STERRY HUNT. 257 lonsumption of 161 bushels of charcoal to the ton. The results of several st. Jiauri(», malyses of the ores of this vicinity, made by me in 1852, are given in ;he Geology of Canada, page 511, and show them to contain more or ess manganese, and a considerable proportion of phosphates. The'ana- ysis of a specimen of grey pig iron made at St. Maurice, in 18G8, gave e the following results for 100 parts. Iron not determined Graphite 2.820 Carbon, combined 1.100 Sulphur 025 Phosphorus 450 Silicon 860 Manganese 1.240 The average produce of the St. Maurice forges is about eight tons of iron daily, which is employed for foundry purposes, and is much esteemed for railway wheels. Some four years ago, a small quantity of Avrought iron was manufactured from it, in a hearth-refinery, but the quaUty of the product was somewhat irregular, and the manufacture was abandoned. It is proposed, in a subsequent report, to give the results of farther studies of these and other irons. South Crosby. A large deposit of magnetic iron ore is found on an island in Mud Lake, on the Rideau Canal, in the township of South Crosby, and not far from Newborough. (See Geology of Canada, page 674.) Considerable quantities of this ore have been mined, and shipped to Pitts- burg, and to Chicago, for use in puddling-furnaces. This ore, however Jcontains, besides an admixture of chloritic matter, a considerable proportion lof titanum, and more or less sulphur in the form of disseminated grains of jpyrites. The specimen selected for examination was frctn a largo block !sent to the Museum of the Survey, by the Messrs. Chaflfoy, some years since. Its analysis showed the presence of considerable am > mts of 3 alumina, magnesia and water, which belong to the intermingled cliloritic I mineral. The iron is calculated as magnetic o.xyd, although a portion, I uncertain in amount, doubtless exists as protoxyd, in combination with the ■ titanic acid, and with silica, besides that wliich enters into the com position I of the sulphuret of iron present. An average sample yielded as follows ; Magnetic oxjd of iron 69.77 = metallic iron 50.23. Titanic acid 9.80 Magnesia 4.50 Alumina 5.65 Silica 7.10 Water 2.45 PhoBpborus 085 .South Crosby, Ontario. Sulphur 1.520 = pyritea 2.85. 100.875 1 258 GEOLOGICAL SURVEY OF DAXADA. [1^ An analysM of another portion of this ore, by Dr. A. A. Hayes of Boston, gave 1.49 of sulphur, 5.04 of silica, 4.42 of magnesia, and 16.45 of titanic! acid. When the pulverized ore is treated with a magnet, it is partiallji purified, the non-magnetic portion retaining the sulphur, and a large par of the titanum. The magnetic portion equalled 74.2 per cent, and con- tained 54.76 per cent of metallic iron and 5.70 of titanic acid. North Crobby. North Groabij. A specimen of iron ore examined from what is said t< be a large deposit on the land of Hon. George W. Allan of Toronto is a bright crystalline magnetite, free from any visible trace of pyrites, and containing but a small amount of sulphur. Its analysis gave Magnetic oxyd of iron 30. 14 = metallic iron C-t.90. Titanic acid 1.03 Uxyd of manganess traces. Alumina 1.33 Lime 82 Magnesia .84 Insoluble 5.25 Pbospborus .COT Sulpbur 120 99.537 The protoxyd and peroxyd of iron in this ore were separately determined, and found to be exactly in the proportions required by theory for the magnetic oxyd. The insoluble residue was chiedy white quartz, with a little black mica and green pyroxene ; it was found in another specimen to equal 10.80 per cent. This is a very fine and valuable ore, and the deposit would seem to be worthy of careful examination. Belmont. Bflmont. The great deposits of iron ore at Belmont have been des- cribed in the Q-cologtj of Canada, page 676 and in the report for 1866, page 100. Since that time, extensive mining operations have there been carried on, and the ore has been shipped to Pittsburg, Pennsylvania. 3Iuch of this was found objectionable, on account of the considerable proportion of sulphur which it contained, but an e.\«avation made in the immediate of Sand-pit ore. the former workings, and on what is called the Sand-pit bed, has yielded "a much purer ore, to which reference is made in Mr; Vennor's Report in this volume, page 161. I obtained, by crushing several fragments of the ore, taken from a pile at the furnace of Messrs. Shoenberger & Blair, in Pittsburg, Pennsylvania, what seemed an average sample. It was reddish from an admixture of hematite, and yielded Magnetic oxyd of iron 72.80 = metallic iron 52.41. Magnesia , 4G Lime 35 '•'■ [14 ^. A. Hayes of Boston, 'ia,ancl 16.45 of titani magnet, it is partiall )hur, and a large j_.' ^•2 per cent, and con titanic acid. I from what is said t, W. Allan of Toronto.^ e trace of pjrites,andi rsis gave ^ = metallic iron O-t.90. )arate] J determined, I bj theory for the kliite quartz, with a 1 another specimen uablo ore, and thej 1. 3nt have been des- e report for 18(36, j as have there been Jnnsjlvania. Much derable proportion 1 the immediate of i bed, has yielded a Jnnor's Report in fragments of the erger & Blair, in '■ It was reddish stallic iron 52.41, 15] REPORT OF DR. T. STERRY HUXT. 259 Carbonate of lime 40 " "magnesia g^ ^"'^ '.'. 3:50 Insoluble j^ -o Phosphorus ^og Sulphur f,,^ 101.142 The analysis of another sample of the Sand-pit ore gave of metallic Iron 48.99, water 3.(35, carbonate of lime 8.03, carbonate of ma-iosia U 48 and insoluble residue 1(3.52. The carbonates were removed, in botli analyses, by acetic acid. The ore contains a considerable admi.xture of a magnesiau silicate decomposable by hydrochloric acid, so that the inso- luble residue contains a proportion of soluble silica, which, in the second analysis here given, was equal to 4.25 per cent. The remainder was a silicate of magnesia, iron and a little lime, approaching to pyroxene in composition. The determinations given in this paragraph a. by my chemical assistant, Mr. (jrordon Broome. Madoc. The Seymour ore-bed in Madoc is described in the Geoloff>/ of Mad„c Canada, page 6 1 0, and is further noticed in the Report for 18(36, pa<'e 98 It was formerly mined and smelted to a small extent, and is a fine -rained magnetite, free from pyrites. The analysis gave me Magnetic oxyd of iron 89.22 = metallic iron 04.23. insoluble j0^2 Phosphorus „p, ^"^p'"^^ ''.".!".!;;".! !o73 99.725 The solution of the ore in hydrochloric acid held neither lime nor man- ganese. The insoluble residue was decomposed by heating with a mixture of auorid and sulphate of ammonium, and gave magnesia 17.15 lime 11 01 protoxyd of iron 11 . 95, silica, by difference, 59 . 89. This is the compo' sition of actmolite, a mineral which is occasionally found in radiatin- masses in the midst of the ore. ° iMcNab. The hematite of McXab is described in the Geolo.pj of Mo-ab C««ai« page 6, 7. It has been mined to some extent, and shipped to tli^ " Umted btates, and was also used in a small amount at the Hull iron- furnace hi 1868, as already described. It is a purplish-red compact or finely crystalline ore, and holds small quantities of silicious matter and of carbonate of l.mo irregularly disseminated. An analysis by me made in 1847, and cited in the Geology of Canada, gave peimyd of iron 84 10 carbonate of lime 8.80, silica 4 . 00. A more complete analysis of anodic^ specimen has since given me as follows : 260 McNab Peroxjd of iron Carbonate of Jime.V.'.', , , ^, " ningnesia insoluble Phosphorus Sujpbur """'"•'" «™VET 0. CV.... [16 Oros Caj,. ®5;^*J="«'«llic iron 59.09. 1.05 7.16 ■ 030 .065 99.125 Gros Cap, Lake Superior Th. A ■ Bay of Seven ■Jslnnci?-. _Pero.X3-,lofiron... ^iisolulile Phosphorus ,'. Sulphur Tliis ore contained n„ 1,^ a„^ « ■ , °''°" Tr"? tf P™ '"■■"»■ " '° '■"*''°' """" ™ "toe ■"".^ ^ ofj;^^ Islands O ' aver rtioh ea,ptie, in,„ lie Bav\rs " '"'r™ ''°»™ "» 'ho Ea„i<l hundred yards frnm •* -^ °^ ^^ven Island^! thr, 1^^'' "»'"e orlaio •: ta't '""' ■"»'' °f '-^0 l\Xd'-^ fr Protoxydofiron. Titanic acid .... insoluble.... '•'•''.'.'.'.■.■.■.';;; si's" =''"''""'■'= ■■■•o'' ss.^o. C.35 90.42 XT '::;;" "r^'^ ^^ne ?:;x%"r" ■' "' '^f""»^^° [16 nietallic iron 59.09. 17] UEPOhT OF DR. T. STERRY HUNT. 261 ^vhich occurs at 'een described hy cimen of the ore, '"ng results. 98.642 'licli was white, as the Rapid ' occurs, a few ibedded in the e e.vception of y the bed and lated at about 'J south, but, t ascertained, Jds imbedded d some iron ?e amount of 00 parts iron 38.70. not deter- irated into remainder d of iron. besides 8.30 of insoluble residue. The magnetic portion, contrary to what might have been expected from the readiness with which it was attracted by the magnet, contained not less than 24.80 per cent of titanic acid. It was nearly free from silicious impurities, and almost wholly soluble in hydrochloric acid. The existence of a highly magnetic compound, con- taining so large a proportion of titanum, is interesting, and the substance deserves further study, — meanwhile, as an iron ore, it must take its place Avith the highly titanic ores, like that of Bay St. Paul, to which reference has already been made. Should it ever be found advantageous to work such ores, the deposit at the Bay of Seven Islands may be made to furnish a very large quantity. IRON SANDS. The silicious sands of most regions contain a greater or less proportion of heavy black grains, which consist chiefly of some ore of iron. The source of these is easily traced to the crystalline rocks which, by their ' ""'< -muu. disintegration, have given rise to the sands, and which, in addition to occa- sional beds or masses of iron ores, generally hold disseminated grains of magnetite, hematite, titanic iron (menaccanite or ilmenite of mineralogists) and more rarely chromic iron ore. In the process of washing earth and sand for gold, diamonds, or tin ore, considerable quantities of these black iron sands are met with, and, from their high specific gravity, remain when the lighter portions are waslicd away. The chromic iron ore is compara- tively rare, and confined to certain districts ; the hematite, with the exception of some crystalline varieties, is generally too soft to resist the abrading forces which have reduced the solid rock to sand, so that the black grains, in most districts, consist chiefly of magnetic and titanic iron ores. In the gold-bearing alluvions of the Chaudiere region in Canada, the sands obtained in washing for gold, when purified as much as possible t.oi«' »iiuvions by washing, were found to hold eighteen per cent, of magnetic iron. The non-magnetic portion was soluble in acids and fused bisulphate of potash, with the exception of 4.8 per cent, of silicious residue, and the solutions contained, besides iron, a considerable proportion of chromium, and 23,15 per cent, of titanic acid, derived from the titanic iron ore, which made up a large portion of the sand. {Geology of Canada, page 520.) The proportion of these ores to the whole mass of ordinary silicious sands is, generally, by no means large, but the action of moving water effects a concentration of the mixture, separating the lighter silicious grains more or less completely from the heavier portions, which consist chiefly of the iron ores, generally with a small quantity of grains of garnet. This separation is eflFccted, on a large scale, by the action of the sea, under the 262 GEOLOGICAL SmVEY OF CANADA. Iron-3aud beds. NewZoalaiicj. "'ostorn iiikr l^vol, ha., ,ome,i„,o, an-an^d "„ ; ! "'■''"""' "' ''^'«'"= "bo" e the .e ■^"Tf ' wi.h the hghto,. mti^lZrZ'n 'T' "•"'=" "^» »-"■ Accumulations of these ir„„ sanS,, '^°''°'''"'' '»-<'V- Jbey are found on the she Jof G et 1.™' f"' ''" """^ -»«™s. Baltic and the Jlcditen-aueaa, and al !?! ' *"« "'" l»*'^ of the 1° some parts of Hindostan and M. ■^'"' "'""""of New Zealand "traced h, .ashing f^nuhe and fotrr '!'° «™"^ "' ^ - - "atives m their primitive furnaces fl!, '''' ''""' ™r'»J""l by the »cale- The iron sands of Ne,v Jah ™""f^"""'> of iron on a sJaU attention from their groat .Z^ItkZ T" ""™'°'' ?"« "' 1^ shore of the northern island fromK'' ^""''''"'S to Hochstetter 180 .ndes, is bordered ,vith a thick ht'T '° ''"""''''' » "^'ta^o f accordmg to different analyses, f™„j:^;;?!'™ »'-'"J. »hich contain . ^" N"--"' America, black ir „ san, , ° ,™ P" «»*■ °f "tanic acid ;n great quantities in the lo"er s j" ' "™^ f'""^' ^^'-^ occur deacnbed, and are met with in sin ■^"™»«. «' "ill be hercafle -uth.west.ard. along the v I e".." te' Tr"' "' ""»- P"'»^' " Tims, a deposit of black sand ,"t (L ^'■>»™"oo and the great lake attracted „„e attention, a w ' e , ".!'' f ..^^^^^Inron, „!ar S f„ ' Lake Er,e this sand is, in some .la !%? '" "'""S ""' ""th shore „ -ere it is said, made more tXulf^f '" "='' ''""""'^ *»' attomp s "Oil .t with an admixture of bo^r Cf ^ ^°'™ ™"' '° '^»"«t it and Zl' t, 'T™"*' N«f„,?co:rt ;j^^^ "'™ '~'«^ » " "last- ■those black sands arp i;i-. • ^' ^'"ano. ;-' of the Dnited'is^ ™^;;' "f J? ™™™ poi.t, along the Mr Home, a steel-maker and cutle o T ". ^^'^''^'' ^^ '"^ letter from 17^.3 that, at that time, the Socie y Cly^'^ '' ^'^^^n, Mareh 3, Manufactures was occupied with tho T ^"'^'^"'•^yonient of Arts and - it was calied. Ah-eid; te i74 " o" T' '^^'^^''"-^ '^^^ - ' Society, had made some un;ucce Ifu e'l '' ^^^"^'''"' '^ «>« ^W' o< th.s magnetic sand, but i^tl veTr'n '' ''^'"•'""- ^'-^ "'^ juantityofit, succeeded, as he to I ' ' ■'^''■"'^' ^''^^ving nrocured n ^^^'^ ^^ ^^« weight of tine i^i^^:';:;;" r:::: ^ ^^ -^-^ ^'- - ^io stems, however, to have'pub- [18 tion occasionally ron sands, alon^ luring the depo- 3 above the sea- hich are seen to I of to-day, nanj countries, borders of the )f New Zealand, of iron ore are mplojed by the iron on a small !ted particular Ilochstetter, i> a distance of ■hich contains, of titanic acid. es. They occur be hereafter 3 points to the 'c groat lakes. 5 near Sarnia, lorth shore of that attempts collect it and ed in a blast- its along the Connecticut, >d were suc- tails relating 'resting that 1 letter from 3 Mr. John , March 3, f Arts and black sand, tlio lloyal tlic nature •rocured a > than one- have'pub- 19] REPORT OF DR. T. STERRY HUNT. 263 lished nothing upon the subject until after Mr. Jared Elliot had made known, twenty years later, by a pamphlet and a letter addressed to the Society of Arts, and subsequently by a letter in reply to Mr. Home's inquiries, EUiot's triau. that he was then making malleable iron from the black sands, in blooms of fifty pounds and upwards, by direct treatment in a common blooraary fire, a process which seems, from his letters, to have been one familiar to him. He describes the ore as yielding 60 per cent, of malleable iron, and as being very abundant, and so free from impurity as to require the addition of cinder or of bog ore. This nianufactnre of iron from the sand had evidently been somewhat developed, for, according to Mr. Elliot, his son had already erected a steel-furnace, before the Act of Parliament was passed prohibiting the manufacture of steel in the colonies. Specimens of the steel there produced wore examined by Mr. Home, and found to be of excellent quality, very tougli, and not at all red-short.* Throughout the essay of Mr. Home the sand-ore is spoken of as coming from Virginia, a name which in the reign of Elizabeth was given to the whole American coast from Canada to Florida, although in 1G43 the name of New England was applied to the region which still bears that name. It appears, hoAvever, that the so-called Virginia sand was from the coast of Connecticut. Mr. Elliot's letter to Mr. Henry Home was dated Kill- ingworth, Oct. 4, 1762. Killingworth is a town in the state of Connecticut, Connecticut, on the shore of Long Island Sound, twenty-five miles east of New Haven, and Avas the residence of the Rev. Jared Elliot, D.D., who was not only a divine, but a physician, and a naturalist of great roput:. It is recorded of him that " some considerations had led him to believe that the black sand, which appears originally on the beach of the sound, might be wrought into iron. He made an experiment upon it in the year 1761, and succeeded. For this discovery he was honored with a medal by the society instituted in London for the Encouragement of Arts, Manufacturers and Com- merce." * Notwithstanding this successful result, the iron sands seem to have been neglected for the Inst century, both in America and in Europe. We read, it is true, that such sands arc treated in open hearths (bloomaries) at Avcllino, near Naples, and within a few years attempts have been made in England to turn to use the iron sands of New Zealand ; but the first successful attempts in this country were on the north shore of the lower * These curioug details are extracted from a rare volume entitled E$tays concerning Iron and Steel, (the first of the three essays being on " The American Sand-Iron,") by Henry Home, London, 1773. 12mo., pp. 223. A copy of this scarce book is in the possession of W, M. B, Hartley, Esq., of New York. * Harber's liislurical Collections qf Conneclicut, page ,')31, The Rct. Jared Elliot, who was a grandson of the celebrated John Elliot of Massachusetts the " Aiiostle of the Indians," died in 1TG3, aged gcreuty-cight years. 264 GEOLOGICAL SURVEY OF CANADA. [20 St. La\Trence. The great deposits of black iron sand on the beach near the mouth of the Moisie River, having attracted attention, various attempts to reduce it -were made. In January, 18G7, Mr. W. M. Molson of Montreal, had the ore successfully treated by the bloomary process, in northern New York, and the result proving satisfactory, several bloomary furnaces were, in 1867, constructed by him at Moisie, and have since been in successful operation. It will hero be well to notice the nature and the composition of the Moisio. iron sand at Moisie, as observed by myself in the summer of 1868. The stratified sands at Moisie, lying about ten feet above high-water mark, penetrated by the roots of small shrubs, and holding marine shells, were observed to be banded by irregular dark colored layers, in which the iron ore predominated. The same thing was afterwards remarked by me in stratified sands at much higher levels in the vicinity. Where these sands form the beach, they are exposed to the action of the waves, which effect a process of concentration, on a grand scale, so that, it is said, after a preva- lence of certain winds, great belts of nearly pure black sand arc exposed along the shore. At the time of my visit trenches were being sunk to a depth of five feet, on the shelving beach, about half-way between high antl low-water mark. The sections presented alternations of nearly pure silicious sand and of black iron sand, the latter in layers of from half an inch to six inches in thickness, often with a small admi.xture of grains of red garnet, which sometimes formed very thin coatings upon the surface of the black layers. One of these latter, six inches in thickness, was taken up by myself, and found to be very pure, as will be seen from its analysis, farther on. It was easy, from these trenches, by means of shovels, to remove, without much admixture, the thicker layers of the moist black sand, which wotild measure from one and a-half to two feet out of the five feet excavated. This material was piled upon the beach, and afterwards carried to the washing-table. The supplies of sand-ore have hitherto been obtained from the deposits of wet sand below high-Avater level. Those at the surface, on the beach, have doubtless been recently moved by the waves, but from the inspection of the layers in the trenches, I was led to the opin')n that thoy were lower strata, similar to tliose seen above the high-water mark, and, like- them, of considerable antiijuity. They were found to contain marine shells in a crumbling and decayed condition. It is said that these mixed sands of the higher levels yield, on an average, by washing, about fifteen per cent, of black iron sand. When this poor sand is spread upon the shore, and exposed to the action of the waves and the tide, it is found to become concentrated through the washing- away of the silicious grains. This process helps us to understand the mode in which the irregular layers of rich iron sand have been formed in the [2a on the beach ttention, various ■• W. M. Molson lary process, in 3voral bloomary md have since iposition of the of 1868. The ;li-wator mark, le shells, were wJiich the iron fked by me in re these sands ', which effect I after a preva- d arc exposed iag sunk to a "cen high and nearly pm-e from half an ' of grai)i3 of I the surface ickness, wa::i een from its means of yers of the •0 feet out hcach, and nd-ore hnve liigh-water ou recently 10 trenches, those seen antiipiity. 1 decayed s yield, on k1. When ion of the > waahini'- the mode led in the 21] REPORT OF DR. T. STERRT HUNT. 265 midst of the deposits of silioious sand, in the strata which are now above the sea-level. The washing of the ore at Moisie, preparatory to smelting, is done upon washing tue a shaking-table, about twenty feet long and four feet wide, with a sloping "'"'^" and somewhat concave bottom. Upon this, by the aid of a gentle current of water, a large part of the lighter grains, chiefly of quartz, are washed away. The specific gravity of the sand, in bulk, was determined by weighing «pooiflc graTity. 100 measured cubic centimeters of it, equivalent to 100 grammes of water ; and the proportion of grains of magnetic ore was also deternlined. Of three specimens from Moisie ; A was an average sample of several hundred tons gathered in the manner just described, preparatory to washing ; B, a portion taken by myself from a layer six inches thick, about three feet below the surface of the beach ; and €, the washed ore, as prepared for the bloomary fire. In this connection are given the results of some similar determinations with iron sandj from other localities. Svecijic gravity. Magnetic. Moisie, A 2.82 46'3 per cent. Moisie, B 2-88 49-3 Moisie, C 2-97 fi2'0 Mingan 2-84 48'3 liergiinla 2-81 34-3 Natasquan — 55-7 Kagaslika — 24°0 13ali3can — 65-0 The specific gravity of the siliclous sand with which these iron sands are associated, was found, when determined in bulk, as above, to be about 2.00. It consists chiefly of quartz, whose real specific gravity is about 2.05 ; that of magnetic iron ore being about 5.18, while the titanic iron ore is about 4.70, and the associated garnet not far from 4.0. The amount of material removed in the process of washing at Moisie is not very great, as may bo seen by comparing the proportion of magnetic grains in A and C, the Moisie sand before and after washing. The latter was found by analysis to contain about 6.5 p. c. of insoluble matter, chiefly silicious sand, the remainder being almost entirely oxyd of iron and titanic acid. The sand of Batiscan, mentioned above, had been purified by washing. Considerable deposits near Ghamplain, contain, according to Dr. Larue, about 10.0 per cent, of magnetic ore, the remainder being chiefly silicious sand. The specimens from Bersimis, Mingan, Natasquan and Kagashka, however, though collected, as I was informed, without washing, compare favorably with those from Moisie, and, with the exception of Bersimis, even 266 «^O.OGI0AL SURv.y OP a.X.D.. BersiiD's, •Bay of Seven JslaiiUs, Jlingan. Q«ebec,who has paid mucra „«„„ It™'''^ '" 1™' UniveSj l«™ce, and colkcted himself ,l! " "■"" '»"'"' "^ «>» lower & locah.^ he has given n,e Sevres i,r'r " '""^ ^'-'"''^' of "hfch ;>.ccu„„lati„„, of sand on fte be, 'f r'"' ^'^'*» *e considelb e mt:?" ™*' *"° '«^- of's :: Tow'' "'r' *- '-'"»- ma«nel,c ore, and separated by a strati „ft«- "T' ^° P""- ««'• of containing very little ir„„. j.,' , ™f 7 ""o"-- mehes of a gray sind fr : rf'f '" "»»» f-' * I't s rr t™ 't' """-««- and h d\ 'f ''™''"« "ontai^ed b„ 343 1' ™ '""' «™. "■" =a».I iind had a speeifle gravity of " si . ,i. ° P" '™'- of magnetic ore :r.h s?:^;';f ih '"^'^'^•^^'tt "»"; '°"- " wese ^vil] be found farther on. " ' ' ^^'^^ ^"^lyses of both ;^ ^;Sd at ^b- :':,:! r: ^r- '» ^'"'■-' - «- -^ ^oisie ^^■B»y »f S«venIsla„ds°to tl°„ tr:?;r«'^.''" "'"* <"»tanee f Z' ft Sf" "'■* '^ »™''°"-l»''o aX :,fr^-' '»'■-■• l'l.o sand from ft|'l'oi- .», ,s said to be fro,n the T.f i V" "'""y^'" "'" !•<> «ive„ ^J'"gan, but is .teeribed as s ret U, ! A 't ""^ •''" ®'' J"!"' Kiver a" f " ;?' Nalasquan and at Ka^ashh a'^-e ,7 "''•■""*• ^''° ''oposils of »"^Hc Mingan, favorably sitS')" ,,e I: "'"' f '° '"^ "'-* An mspeetion of the iron smdrr f = »' ''osscis. "ontioned, shows that they all »v ,'" ."° ™"»"^ 'oealities .bove irrr "'r' =-■"'•-'»' os,^;t "r-"" »' "■»■■• »"- '•"t, m the specimens sub. t^ '!vr •' "^'"'^ '^ '^'^'^y ^i^an^ '>7 « n^agnet, tbe poles of wbichCr, ^T' °'"''''"^ ^^^''^ ^a^en un anaiy.od separately, the el 7,:°'" ■ V"" "'•" "^ ^I'^^ZZ » ..ch as is well k„o,v„, disso^ '^ t:';*^^^^ ':^''-'"»« "" and, n th certan, preeaiuions, may be "l"n ,? ,"'™ "'"' S™' facditv "ta,.,c n-on ore. For this p,,rpofc tUt ""'"S™'»'j' employed to dissolv'e '■-■y fi..oly powdered and ift, 1 ;,""'''""" P»«'»". '»'"'« eon -■«'■• of Mroehlorie acid of''^ ^ effi! '; If^Vr!' "'""," '™ «"« »" biavit;- I.Iil, or thereabouts, for [22 lebted for all of ival Universitj, )f the lower St. •simis, of which he considerable hree feet above 30 per cent, of of a grajr sand with consider- seon, the sand magnetic ore, id, however, a lalyses of both 'lat of Moisie, distance from 'he sand from will be given m lliver, at mco of three ' deposits of y extensive, ties above on, a small The latter '>y careful separate ai'Iy pure taiiic iron, 5f silicious eparating taken up and this s far as lod were 'I'ie acid, fiicilitv, 'lissolvo ,1,' been lines its II ts, for 23] REPORT OF DR. T. STERRY HUNT. 267 several hours, or until the undissolved residue is no longer black, but grayish or brownish in color. If the process has been conducted with care, (.homicai and without over-heating, the whole of the iron, and all of the titanic acid which -was combined with it, will be found in solution, and may be sepa- rated by the ordinary methods. The residue, apparently, contains little else than graias of quartz, with a small proportion of garnet. The finely pulverized ore may also be fused with bisulphate of soda, a process which is more expeditious, and yields equally good results with the last. Moisie. — A specimen of unwashed black sand from Moisie, holding MoiMujauii. 49.1 per cent of magnetic grains, was decomposed by digestion with hydro- chloric acid, and the residue fused Avith bisulphate of soda. The titanic acid having been thrown down, by boiling, from the united solutions, the iron was directly determined, the other bases being neglected in this partial analysis, which gave me the following results : I. Protosyd of iron 70.10 = metallic iron D5.23 Titanic acid 16.00 Insoluble, chiefly quartz 5.92 92.02 A part of the iron in these ores is in a higher state of oxydation than here indicated, but the determination of the degree of oxydation of the iron in titanic ores is difficult, and, as even the magnetic portion of the sands contains some titanic acid, it is thought advisable, in the present analyses, to represent the whole of the iron in these ores as protoxyd, giving, at the same time, the amount of metallic iron, and, in the case ot the magnetic portions, the magnetic oxyd corresponding thereto. In the non-magnetic portion of the Bersimis sand, however, as will be seen, the proportions of the two oxyds of iron were determined. The magnetic grains liaving been removed from the above sample of Moisie ore, the non- magnetic portion gave 58.20 of proto.xyd of iron, iO.l-i of titanic acid, and 6.14 of insoluble residue. Further and more complete analyses were subsequently made of the washed ore from the Moisie iron-works, which, as already stated, contained ")2.0 per cent, of magnetic grains. These were analyzed separately, (II) while the non-magnetic portion gave mo the results under III. Sulphur and phosfihorus are present in this sand in very small (piantities, the determinations of Mr. Broome giving for the washed mixed ore .070 per cent, of sulphur and .007 of phosphorus. 268 "'^^'^^^^^^ «^«VEr OF CAN,,,. Moisie sand. [24 Protoxydof iron.. "• Hi Titanic acid... 85.79 gg gg ^ ■*• Oxydofmauganele!! "■'« 28.95 J«t^ Lime .40 J jQ ^^•S'^ Insoluble "._' •• .90 ^g 1.95 o'-r • ■ _^^^ 5. .S3 Magnetic ox^d of iron.. "^ -!!f """^ Aletallic iron... ^2.68 " 66.73 43V" *'^-^'» 55.27 P'-otoxyd of iron I^'- Titanic acid 85. 5G OiJ'J of mangauesV." ". ] S**^ Lime undet. Wagnesia ' _'_" traces. Insoluble 3 85 go p ] Magnetic osyd of iron... ' - Metallic iron 92.44 66.56 . ■''J'-O'Woric aoiJ, „u,°of contacw, „ "'"™^ '""^ ™ '"i^lrnl i Piotoxjd of iron V. Peroxvd of iron . . .*.'. 24.66 Titanic acid ."'"_" 22.24 Oxyd of manganese 26.95 I-'n'e ■■' i.io Magnesia '' 1.12 Insoluble .72 23.80 IOC. 59 Metallic iron — 34.94 N#W [24 III. 1 A. .38 71.08 95 16.55 10 95 5.35 55.27 as magnetic oxyd, ippose II and III ■ A, which agree.s icribeJ, contained s portion is given ^nietic oxjd, is IS dissolved in s of protoxyd Ijsis gave mo 25] REPORT OF DR. T. STERRY HDXT. 269 Mingan. — The iron sand from the mouth of the St. John river, at Mingan, contained 48.3 per cent, of magnetic grains, whose analysis is *''"-''" ^''"''• given under VI, while that of the non-magnetic portion of the ore is found under VII. VI. Protoxyd of iron 80.46 Titanic acid C.50 Oxyd of manganese 52 Lime 75 Magnesia .70 Insoluble 4.20 93.13 Magnetic osyd of iron 80.92 Metallic iron C5.53 vir. 46.31 31.60 1.35 1.06 ..•50 15.50 9C.32 36.00 Tlic sum of the analysis VI, if the iron be estimated as magnetic oxyd, is 09.59. In the above analyses of the iron sands it will be remarked that the magnetic portion retains a little adherent silicious matter, and small amounts of titanium, both of which vary in the sands from different localities, although the separation by means of the magnet was in all cases effected with the same precautions. Observations and experiments on other samples of these sands go to show that different layers from the same locality vary, not only in the proportion of silicious sand, but in the relative proportions of magnetic and titanic ores and of garnet. This might be expected when we consider that the differences in density between each of these consti- tuents of the sand, should, under the influence of moving water, lead to their partial separation from each other. A specimen of iron sand from Quogue, on the south side of Long Island, near New York, where these sands arc about to be employed for the manu- factui-* of steel, closely resembled those of Bersimis, and contained ol per cent, of magnetic grains. The unpurified ore, which was mingled with a considerable amount of quartz sand, and some garnet, amounting together to about 17 per cent., gave by analysis about 40 per cent, of iron, and 15 I'cr cent, of titanium, besides a proportion of manganese greater than the iron sands from the lower St. Lawrence. VaryniK oom- lic-fitioti. OX THE MANUFACTURE OF IRON AND STEEL BY DIRECT METHODS. Although by far the greater part of the wrought iron and steel now used in the arts is made from cast iron produced in the blast-furaaco, 270 OEOLOOICL SUKVEY OP c.VxD.. [26 Furnaces Catalan torg, even ^"eZ »-er„CitTatoI,l" r,'!'' '" '"'^ '™« "■""oable iron '„,, pr.«.„, »c latter, m fact, date only f,„„ , „ ' ■"'« ■"Mufaotare and us. of "»'.ve, of I„di„, ji„„,^ ^ ^'O" J omparativei^, recent period The 2. ; . ""•■"" '''^"ieta of I„,li., ■ '""•''"'° »'»'« in snni .w nt. "7T "*'' "'"^ ^™" i» =^d '':;"""'"■'''">«'' i"to steel "enty to forty pounds of iron in a /,„ ,f , '° ""' "">'» U'an from "n,andwithagreat™te„for"a„UfT *" '""" "' tlno. or Z Tf "{ P-lvmed, or the gratas of' ''™'"- ^'« "">' «"Kve o e, ^™ls of eortain districts, areCedtirV™ °'"'""'^'' '^ "^in-Io nveision of ores mto malleable iron H « f "'"'""^^ ^^^ «»e direct torsican and Catalan forces Lr ^"' ^"^^^"^ '" Europe areZ f^»-«ace, and the Gorman ^jSi- / ^''"^■'*" '^'^^'"ary for^e thoO ? nri I,; 7 „ " ^'^^ickofcn or fiifri. n^ -^'o^j "le Osmund readier is referred to D Pet ' T'^' '" '" ^^^^^^ ^^ -'-J^- 1. ml ^n^J^teel. Inasmuch J^oJve;^^^^^^ "'^'•^' -^ ^he metallu^g! o 'i '' are still largely used on , "''.^"'"ac<^^i'eiated to the Gorrr,! n " iMn ;.v, . •'^ "^^ ""s cont ncnt i.>,l v, • ^c'TOian bloomarv a We nnportance to Canada it ,.-ii i ' P^"'^'"'^'^<^ to become ofonn^-i from thepr^inoo 'c, Xf^ '? t° -"-'"'t" "■"'■»«« ^ f«r"e so eal, , Olopartment of AriO-p^ „ "^^ ^^''^ -^^-f^nch Pyrenoo, , " ^'^ liv.?.-r.„a ) -^>i<-ge; the ore jrrenerilKr „o i • /'*^"^cs, however, '' ^«^alan forge consists of a rec^an^l^ , T'""'' '^ ^'''^'<^- octangular hearth, constructed chiefly [26 I malleable iron, and nthout tlie previous lufacture and use of ecent period. The 'me parts of Africa, :allic state in small malleable iron thus 'actured into steel ; ot more than from T of three or four 'le rich native ores, I by washing the ifill furnaces, until Somewhat similar 1 in various coun- still followed, and cos for the direct 1 Europe are the 'ge, the Osmund ace, which latter blast-furnace, of I For a detailed I'l^ing them, the »^'Sy of iron Jrman bloomary "'■ of consider- some points 27] REPORT OF DR. T: STERRY HUNT. 271 >me fly ''gc, so called y much used, cut of Ariegc, 'lucing' oS'jO le remainder lice probably " the Italian the province iilar iron ore ^, howevci', irnaces is a of iron, and Quebec, cted chiefly of heavy iron plates, which, in the largest size, is about forty by thirty- two inches, and from twenty-four to twenty-seven inches deep, or from fourteen to fifteen inches below the twyer. In some districts, however, furnaces of not more than ona-half these dimensions are built. The pres- sure of the blast employed does not exceed li or If inches of mercury, and the twycr is directed downwards, at an angle of thirty or forty degrees. The wall facing the twyer, slopes outward toAvards the top, and in working, the greater part of the charge of ore is heaped against it, and occupies from one-third to one-half of the cavity of the furnace, the remaining space being filled with ignited charcoal. The ore is previously broken so that the large lumps are not more than two inches in diameter, MoJoof^oT* while from one-third to one-half of the material will pass through a screen, the bars of which are four-tenths of an inch apart. This finer ore is thrown on the surface of the fire, from time to time, during the operation, which is conducted with many precautions as to regulating the blast, stirring, supplying the fine ore and coal. At the end of six hours, in the ordinary routine, there is withdrawn from the bottom of the furnace an agglomerated mass of reduced but unmeltcd iron, which is then forged into blooms or bars. The operation, lasting six hours, consumes, in one of the larger sized forges, about 9i cwt. of ore and 10^ cwt. of charcoal, and yields 3 cwt. of bar iron. According to another calculation, there are recjuired for the production of 100 pounds of iron, 340 pounds of charcoal and 312 pounds of an ore containing from 45 to 48 per cent, of iron. Of this about seven-tenths are obtained in the metallic state, the remaining three-tenths passing into the slag. 100 pounds of ore yield 31 pounds of bar iron, and 41 pounds of slags, which are dark-colored basic silicates, very rich in oxyd of iron. The Corsican forge is a more primitive form of furnace than the Catalan, corsican forge, and without interest, except so far as it belongs to the history of iron- working. It is said to have consumed more than 800 pounds of charcoal for the production of 100 pounds of iron. Some few of these forges were still in operation in Corsica forty years since. Another form of furnace, described by Dr. Percy under the name of the Osmund furnace, was used during the last century in Norway and Sweden, o-munarurnacfc It was a rude hearth, with walls around it, and an opening in one of the side near the tap-hole, which was built up with stones, and taken down when it was required to extract the loup or mass of reduced iron. This furnace was not capable of yielding more than IJ ton.s of iron in a week, but is still used in Finland, and it is mentioned as a curious fact, that certain bog ores which contain so much phosphorus as to yield l)ut a poor and hot-short iron by treatment in the blast-furnace, and subseiiucnt dccar- buration, afford a good malleable iron when reduced by the direct method, 272 OEOLOOICAL SrKVEV OF CANADA. Improveil c»t::. J«n, or Oeneusu ;° "le Osmund fur„ace ; a «3„1, „., • , "^ f"".ace, eacape, redaction '^1. J'"? '"'" "'' '™. » "'= b furnace. . '!>« lower temperature of the OaJuS wh.eh was a vertical chamber, Lnn tic J-^^^^^^^^^^^^^ -thin and above' a Side-door, and a g.-ating at tr^t 1 "f "'''' *'^^ ^^i^^ney, and haviV secular oxyd containing°68 n r cen of ''" '''^ S^'-^""- «- ore, ^ J -^^t, winch roasted it, el-pollin ' ^ n ''"' ''"' ^''"^^ '^^ <^-^Posed to he ; ;V'-^^^ for son. ti^ ' I 3 :^ ,Pf '- «^ -Iphur. ifter be ! an coarse powder, it was spread outoVf ° '" '^'■'^■^" '"^^ '^^^^ lu^pl ^^"'"^ ^^-J'-^J' the bottom of tSe rev 1 . •" "V '^^^^ '^ '^^oken chared. ;overed, and was here exp ed « t™, T '^'^^"^ ^"^ P-viousi; ' f-n the forge, during the w t t m o ". ? '^"* ^^ *^'^ -^^« Aam In this operation the bed of I! , ^'''^'"= * '^'^arge in tho htfPr or twelve per cent, of ts w'L T "" """"^^' ''^"'J ^he o ,o i -»'.ouv,„e!,arc„alJ„p,,„,3;ff " '«"ty-fo»r houra, with gj "■:«"<;.■ J-ield. Separa e f„, ° "' '° ''f "^ "^ '™"' "ud a sotnerta ;■"' 'l'«o works, for reheati, j,.!!" f V»"3'™cled i„ eouuect on the waste heat from these ,v, °aL , '" ''° •''•'"™ »»« «>'<> blooms and above e.plai„ed, "' "''" '"'"'''^''' ""«.in,, reverberator; "t """"""■"• , 0"« »f the Catalan for.-es ,,i.l, ., • °f -^- ;%s, thirty boats ofLr i ;T ""'•-•<»'™«"'^^ yW* in a week lioat, of 05.30 kilogramme, of o I '""'■'«'' ""'"n^ption, for eacb o -ought,ro„ scrap, Z S. ^f" ^ ''f^""' .•"Powder .," W^^ramme,, of bar ir„„. This i e ,ual t^' '""', " ^'*' "' 1^^.00 fo" l;ours, with a consumption „ 070! ,° , '^'"'* "f'™" "' '"'onty. over, to be noticed that about 09' 7 ' "'' ""^ "'""■"""I- I' is, liow- ™ a*Iod in tbe conditi „ tZl "'"''' P-^'-'- or 349 p„„ I consume eompuativelj little cSiri''""''' *»^ "^tiuS "ol P'o.luct,o„, from .bo ore, of imZZr^'', ", "°"'^ ''!""' '° «- pounds of „.„„, ,,l„„|, ,s at the rate of 50 [2 due to the fact tha 5 iron, in the blast •e of the Osmund iced in the province n of the waste heat, luce, the ore before Ided reverberator^' me from the forge, within and above limney, and having grating, the ore, a ind exposed to tlie Imr. After being wn into water, bj n into small lumps f broken charcoal, i previously been f the waste flame fge in the latter. i the ore lost ten ed. Some scrap.s lalf-reduced ore, gh the charging- waj, five heats, urs, with great and a somewhat in connection ito blooms, and erberatories, as icids in a week 'tion, for each owder, 31.75 eld of 143.00 ■on in twentj- It is, how- t" 349 pounds, orking would ance for this, O'lual to the le rate of 50 !9] REPORT OF DR. T. 8TERRY HUNT. 273 pounds of iron for 100 pounds of charcoal consumed, and is about the result obtained with the American bloomaries, to be noticed farther on ; while the proportion obtained with the unimproved Catalan forge, des- cribed above, is only at the rate of 30 pounds of iron to 100 pounds of charcoal. Mention has already been made of the German high-bloomary furnace, or Stuckofen, which is of no particular interest in this connection, and is not to be confounded with another furnace known simply as the German bloomary. This was formerly used in Silesia and the Palatinate, and is described at some length in the classic work of Karstcn, written a little more than half a century since (181G), but is dismissed with a few words in Bruno Kerl's treatise on metallurgy, published in 18G4 (Ruttenkumh, iii, 427), from which its use would seem to be nearly or quite abandoned in Germany. According to Karsten the German bloomary consisted of an iron pot, or a box of iron plates, in either case lined with refractory bricks, and having an internal diameter of from fourteen to twenty-one inches, and the same depth, the dimensions varying with the fusibility of the ore, the force of the blast and tlie (juality of the coal. The twyer was horizontal ; the furnace having been filled and heaped up with bumini; charcoal, the ore was thrown ujjon the fire by shovels-full at a time ; thi^ process was continued, the supply of fuel being renewed, until a loup n; sufficient size had been formed at the bottom of the hearth, as already described in the Catalan method. When the blast is too intense, or t'.ic coal very dense, it may happen that the reduced iron becomes carburettt-d to such an extent a« to produce steel-like iron, or even molten cast-iron, instead of a loup of soft malleable iron. A similar state of things some- times occurs in the Catalan forge, and is occasionally taken advantage of to obtain an imperfect kind of steel. From the above (lescri})tion it will be seen German bloomary differs from that by the Catalan forge, in tin' fact that, in the latter, the greater part of the charge of ore is placed, at the com- mencement of the operation, in a coarsely broken state, on the sloping wall of the furnace, opposite to the twyer, while the remaining portion is subsequently projected, in a more finely divided comlition, upon the surface of the fire. h\ the German method, on the contrary, the wii ile of the ore is reduced to this finer condition, and is added by small portions, a plan which dispenses with the charging of the furnace after each operation, as in the Catalan method, and permits of a continuous working, interrupted only by the withdrawing of the loups from time to time. The German bloomary, in an improved form, is extensively used for the reduction of iron ores in the United States, where it is known by the name of the bloomary fire, the Jersey forge, or the Champlain forge, and is also frequently called (Jortnan blom- ary. that the method bv the uistinKuish..k iioiu Catalau. 274 OEOLOOICAL SURVEY OF CAX.DA. proceeding to describe in detail 1 a ""^"'^^ '^ ^'^ ^'^''ked. Before to notice son^e of the J.aTJZ^^^^^^ ^^e, it .,„ ..t; f'-om Its ores, and to point oS the t V^' ''''^"'^' '^ ^•^'•'^etin. iro "sod with advantage. *''' ''"^^'^^'^'^^ "nder which they may b" — .. .X:^t t :i:r r ^ ^ ^^-^ — ^^ o.en or . ^Hheore is e«.eted by a SI ^^^^^^'^ ^f the foreign .Itll: ; ' ^oreover, that certain mvnrCti^^^ f/V" '^ ''''^^''' ^^^'-^ '• ^>^th the n-on at Jiigher temperatu el' n ^ ^ '^^ ^"•^^^"^^^^ ^'^"d unite "-educed state, at the lower ^rof , ! '"'"' °^ '^^ "- «^ags, in a^ tra ion of that Juis been given a ovl ?'" ''»'• ^ «^'''^^'"'^ iHu -^^ts use in Finland. C^:^2:r^;i^ '' ''' ^^-"'» ^-at "^ ««^e regions, and with certain ores h ''"' *^^*''^ «P""on that T" f ^'"tageous than the uJof ,; t 'T ^"°"^ ^-^' P-hap finery hearth. This, however, wa ^ i '"^"'■"''^'^*^ '^^"^'-^^e^ vvith the "meantime, great improvements have b. ' T"'"''^ ^'"^^' ^»d, in ";n, as well as in iLldlin. t Z.. .'"''^' ^" *'^^ ^anufactu e of ca o all these facts, and of the°;re oZ rl'"° "^^ ^^'°--^'^'- ^ - w < ay Dr. Percy observes (in 1 ) t^.^, ""^^"^^^ «* the present tively fow localities in Eurone .1 f ''° '^'^'^ ^"^7 be comnari conducted with profit, t Tun H '''''' (Catalan) forges can ^ -san.u.od suitable for^ir:^^^^ '^'-"^^"^ i? nriro <^atalan process may hold its <.round I . '"accessible to railways the -- unprotected b^ high ratefof et- h^T^l^^^ ^" ^^^ o ^petition with iron smelted and manS' n '^'^' ^^^^'^'^^tances, from advantages are that the outlay andTolt '' ^^ '"°^^^" P'-«<^esses. It. -considerable, and the consum';Ln o^c frclT ''''''''' ''' ^ ^-=- ar -ilie German bloomary process w., , J,"^ America early in the Jt intZ l^ll^ i"^-^-ed into North xVwJerseyandPennsylvaniainl856 Les? VT' ^" <^P*^'-ation in ^jnde, mentions one as having b en ± r-^' ^^"^ ^'"^^ ^^-^"/«^^«mV t t W n^'"^ ^""' P^^^'^P^' bloom rielt^^^^^^ " '''•'' ^"^ -««-'• - the Walloon method, which was us!d n .1 ^''^'''^'^^<^rsion of pig.iron by ^t IS evident, from facts cited a eadl ' 0/'°'''" ^* ^" ^arly date • bu^ vemed iron ores in the Ger^ ItrnTv f ' ""^ '^^ ^''^^^^"^ '^'h- Connecticut as early as 1761. It was Ir.h kV'^I''''^^^^ P^-a^tised in immigrants which led to the use of t' T^ ^ ^' ^^' ^°^^"S ^^ German ^orge, which, so far as I ca "rn i ttr"" '''''' '''' ^'^ ^"aln -nd eastern parts of the United SteT'-'* '''''' '' *^^ "-thern '• ^^"^"3 improvements have Bloomarie8 in America. [30 ■n, it is distinct in worked. Before ire, it will be well )f extracting iron lich thej may be )ces3 is often of a e foreign matters complete fusion ; ?duced and unite y the slags, in an A striking ilkis- 3smund furnace, ' the opinion that 33 was, perhaps, mbined with the 3e, and, in the nufacture of cast metal. In view •n at the present ilj be compara- [ forges can be in rich iron railways, the ocalities where mstances, from processes. It^ or a forge are ely small." — d into North oj)eration in ''anufacturers'' id another in pig-iron by y date ; but ment of pul- practisod in of German the Catalan he northern ments have 81] REPORT OF I)R. T. STERRT HUNT. 275 been, from time to time, made in the construction of the furnaces, the most important of which has been the introduction of the hot blast. Favored by supplies of rich ores, and protected, to a certain extent, from [•',','","js{ate*. foreign competition, by duties on imported iron, the manufacture of iron by this method has been widely extended over the United States, and has assumed considerable importance. In the districts where it was first introduced, including northern New Jersey and the adjacent portions of New York and Pennsylvania, the bloomary process is falling into disuse, since wood has become scarce, and extensive workings of coal in the vicinity, with the great facilities for transportation, have rendered it more profitable to treat the ores in the blast-furnace than in the bloomary fire. In northern New York, on the contrary, the use of the direct process appears to have considerably extended during the past few years. The works for producing iron directly from the ores, by the present method, are known in the United States as forges or blooniarics, and sometimes consist of twenty forge-fires or furnaces, but in many cases of not more than two or three. According to the report prepared by Mr. Charles E. Smith, for the Iron MamifadurerU Guide (page 760), and published by authority of the American Iron Association, there were, in the year 1856, produced directly from the ore, 28,033 tons of malleable iron, from 203 forge-fires. Of these, 42 were in New York, 48 in New Jersey, 36 in North Carolina, 14 in Alabama, and 50 in Tennessee. There were besides, at that time, 35 abandoned fires, of which not less than 29 were in New Jersey. The average production from each forge- fire was thus 141 tons. Since that time 1 have no means of knowing the conditio", of this manufacture in the south and west. In New Jersey, for reasons already given, the direct method is almost abandoned, while in northern New York, on the contrary, it has greatly increased. Instead of n.-w vork the 42 fires reported in 1856, there were, in 1867, according to the Iron and Steel Association Bulletin, 136 fii-es in activity in Essex and Clinton counties, the principal seats of this industry. The aggregate product of these forges was supposed by a competent authority, in 1868, to be nearly 40,000 tons of malleable iron, a large portion of which is consumed at Pittsburg for the manufacture of steel by cementation, a process for which this iron is eminently fitted, and for which that reduced from the ore of the Palmer ore-bed, near Keeseville, is especially prized. Two establish- ments in the neighborhood work the ore of this deposit ; one, that of Messrs. Rogers, of Ausable Forks, had 21 fires, and the other, that of the Peru Company, of Clintonville, 18 fires, in 1868. The direct method of reduction cannot be applied to poor ores, which, to yield good results in the German or Catalan forge, should not contain much less than 50 per cent, of iron, while much richer ores are to be pre- 276 GEOLOGICAL SURVEY OF CANADA. [32 ferred. Some of the iron ores of North America consist of an aggregate of crystalline grains of magnetic oxyd, mingled with so large a proportion of calcareous or silicions matter as render them unfit for the bloomary fire, without purification. This is generally eft'eoted by crushing and wash- ing, after a previous partial calcination, and leaves the ore in a coarsely granular state, which would not be adapted to the Catalan, although well suited to the German or American method. This condition of things is I'aimer oi.-bcii. illustrated by the ore of the famous Palmer bed, just mentioned. I was informed at the works of Messrs. Rogers, that from four to five tons o^ the average cruile ore were re(juired to make a ton of blooms. The , , as raised from the mine, is chiefly magnetite, Avith grains of white quartz, and, in some portiivis, of flesh-red feldspar. It is slightly roasted, to render it fria- ble, then stumped and passed through screens with openings of about one eighth of an inch, and purified by washing. Two tons of the washed ore were required to make a ton of lilooms. 1 took what seemed an average sam- ple of the crushed ore from the stamps, and having further reduced it so that it would pass through tbe mcshoa of a sieve having sixteen holes lo the linear inch, carefully separated the magnetic from the non-magnetic part, which contained a proportion of grains of siieoular iron ore, but was chiefly (piartz. The magnetic portion e([ualled 45 per cent, of the whole. A sample of the dressed ore, such as supjiliod to the bloomaries, was treated in the same manner, by fui-ther crushing, and separation by the mag- net, and contained ti4 per cent, of magnetic ore ; the non-magnetic portion, besides silicious matters, holding a considerable proportion of grains of specular iron, which would probably raise the amount of oxyd of iron in this samiile of the water-dressed ore to about 85 per cent., or a little over 00 percent, of metallic iron. In other districts of northern New Vork,as in the vicinity of Port Henry, the cruile ore^ are richer than those just mention' d, and often contain very little extranoinis matter, so that the ojierat' .n of washing may sometimes be disjiensed with. At the Mew Rus- sia ibrge, in Moriah, the ore, which is mingled with a little ipiartz, is roasted in j)iles. with wood, during two or three days, then crushed and treated as above described. Two tons of the crude ore yield one and a half 0*" dressed ore, which is caIo\dated to give one ton of blooms. The wash- ing process removes not only the foreign matters, but a jiortion of fine ore, which is lost, and may be seen accumulated in the vicinity of the washing- tables. The bloomers, as the iron-makers arc called, object to this fine ore. as being unfit for use, but it will be seen further on that this prejudice i.-< withmit foundation, and that the finer grains can be used with advantage, though they are now rejected, and cimsiderabte loss is thereby incurred. The magnetic ores of Lake Chamjilain are exported to Vermont, wliere, for several years, a few bloomaries have boon supplied with iron ore from [ )f an aggregate ■ge a proportion e bloom ary fire, ling and wash- re in a coarsely 1, although well tion of things is tiuned. I was five tons of the 5. Tho ( , as e quartz, and, in render it fria- igs of about one ivashod ore were in average sam- !r reduced it so sixteen holes to e non-magnetic on ore, but was t. of the whole. )loomaries, was tion by the mag- netic portion, o[' grains of oxyd of iron or a little over New Vork, as lan those just r, so that the the New Rus- tle (luartz. is 1 crushed and )i;e and a halt' I. The wash- on of fine ore , r the washhig- this fine ore, prejudice is h advantage, »y incurred, 'uiont, whore. liron ore from 33] REPORT OF DR. T. STERRY HUNT. 277 the west aide of the lake. Three forge-fires were, in 1868, in operation vomont at Salisbury, and three at East Middlebury, Vermont, five miles from the *°'''*''' Middlebury station on the Rutland and Burlington Railway. The ore for this purpose is brought by water from Port Henry or Port Kent to Bur- lington, and thence by rail to I\Iiddlebury station. This is brought partly in lumps, which are crushed and washed at the forge, and partly dressed to a high degree of purity, and ready for use. Overman is, so far as I am aware, the only writer who has given any account of the American bloomary ]irocess. In his Treatise oti 3Ietallur(/if (sixth edition, 18(J8, page 5-il), will be found a description, accompanied by figures. My own observation, as here given, have enabled me to verify the general correctness and trustworthiness of Overman's statements with regard to this subject. The bloomary hearths or furnaces in ditferent localities exhibit some lit- jnoomary •' . . _ lifartlij. tie variations in size and in the details of their arrangements. The size of the hearth varies from twenty-seven by tb.irty to tv;c:ity-c-lght by thirty-two inciies, and the height, from twenty to twenty-five inches above the twyer, and from eight to fourteen inches below. The sides are made of heavy cast iron plate, and the bottom, although often of beaten earth or cinders, is, in the best constructed hearths, also of iron, made hollow, and kej)! cool by a current of water, which is made to circulate through it. In the East Middlebury forges this bottom-plate is four inches thick, and lias within it a hollow space of two inches. The side-plates, which slope gently inwards, in descending, and rest on ledges on the bottom-plate, are one and a-quarter inches thick, A water-box, measuring twelve by eight inches, is let into the ♦■ vyer-jjlate, and a stream of cold water circulates through this box, and through the bottom-plate, as well as around the twyor. The length of the hearth, from the twyer-plate to that opposite, is twenty-four and a half inclies, and the breadth from front to rear is twonty-nim^ inches. The twyer enters twelve inches above the bottom, and is inclined downwards at such an angle that the blast would strike the middle of the hearth. The (ipening of the twyer has the form of the segment of a circle, and is one inch high by one and three-quarter inches wide. In front of the furnace, at sixteen inches from the bottom, is jtlaced a fiat iron hearth, eighteen inches wide. The side-plate beneath it is provided with a tap-hole, through which the melted slag or cinder may be drawn oif, from time to time. The iron plates used in the construction of these furnaces last for two years. In the furnaces used at the New-Russia works in M(n'iali, already nientioneil, the iron bottom-plate is not made use of. the bed consisting of beaten-down earth or ashes. These fu/naces have a depth of twenty-four inches, and ineasuro twenty by thirty-two inches nt the top, but arc somewhat smaller towards the bottom ; the twyer, in those, enters one of the narrower 278 GEOLOGICAL SURVEY OF CANADA. [34 AVorking of ■blooimiricf. sides of the rectangle. While these are somenvhat smaller than the forges at East Middlebury, those lately constructed at Moisie are somewhat larger, measuring thirty by thirty-two inches, the bottom-plate being fourteen inches below the twyer, which is placed nearly horizontal, but of the same size as that described above. The blast employed in the American bloomaries has a pressure of from li to If pounds, and is heated by passing through a series of cast-iron tubes, placed in an upper chamber, above the furnace. These are in the form of inverted siphons, each limb being about seven feet in length, their exterior diameter seven, and their interior diameter five inches. At the East Middlebury forges the air is made to pass successively through three such tubes, heated to dull redness, and attains a temperature estima- ted at from 550*^ to GOO'' Fahrenheit. The use of the hot blast hastens the operation, and enables the workmen to produce a larger quantity of iron in a given time, than with the cold blast, while, at the same time, it effects a considerable saving in fuel. It is said that where 240 bushels of charcoal will produce a ton of iron with the hot blast, 300 bushels of the same coal would bo consumed if the cold blast were used. The quality of the metal is supposed to be deteriorated if too hot a blast is used. With judicious management, however, the use of the hot blast offers great advantages over the cold blast, and has been very generally adopted in the American bloomaries. The working of these furnaces is conducted in the following manner : The fire being kept active, and the furnace heaped with coal, the coarsely pul- verized ore is scattered, at short intervals, upon the top of the burning fuel, and in its passage downwards is reduced to the metallic state, but reaches* the bottom without being melted, and there accumulates, the grains agglo- merating into an irregular mass or loup, as it is termed, while the earthy mat- ters form a liquid slag or cinder, which lies around and above it, and is drawn off from time to time through the openings in the front plate. At the end of two or three hours, or when a sufficiently large loup is formed, this is lifted by means of a bar, from the bottom, brought before the twyor for a few minutes, to give it a greater heat, and then carried to the hammer, whore it is wrought into a bloom ; the bloomary fire itself being generally used for re-heating. This operation concluded, the addition of ore to the fire is resumed, and the production of iron is thus kept up, with but little interruption. In this way, a skilled workman will, with a largo sized fur- nace, bring out a loup of 300 pounds every three hours, thus making the produce of the day of twenty-four hours, 2,400 pounds of blooms ; in some cases, it is said, 1,500 pounds, and even more, are produced by twelve hours working. In this connection may bo mentioned an arrangement, described and ml of air [34 ban the forges aewhat larger, burteen inches 6 same size as issure of from !S of cast-irou lese are in the set in length, /e inches. At ively through rature estiraa- 1 blast hastens aantity of iron QC, it effects a ils of charcoal the same coal 7 of the metal Viih judicious vantages over the American Banner: The coarsely pul- jurning fuel, but reachea grams agglo- VQ earthy mat- it, and is late. At p is formed, e the twyor the hammer, g generally ore to the th but little ;o sized fur- making the IS ; in some wclvc hours Bcribcd and 35] REPORT OF DR. T. STERRY HUNT. 279 figured by Overman, in which *:hc waste heat from the forge, (or rather from two forges united,) passes into an oven or stove, placed at a level above the bloomary-fire, and there serves to re-heat the blooms, when it is required to draw them out into bars. A set of small blast-pipes, placed just above the forge, serves to heat a portion of air, which is led into the oven, and there burns any escaping carbonic oxyd gas. The air and gases from the re-heating oven are afterwards employed to heat the blast for the bloomary hearth, in the usual manner. I have not seen this arrangement in operation. The following observations will serve to give some notions of the working of the bloomary process in the United States. At the Ausable works, as already stated, the somewhat lean ores are dressed so as to yield about fifty per cent, of iron, two tons of ore being required for one ton of blooms, while at the New Russia forges, in Moriah, near Port Henry, where a nearly pure magnetite is employed, three tons of the dressed ore are stated to yield two tons of blooms. When it is considered that perfectly pure magnetite contains only 72.0 per cent, of iron, this proportion of GU.G pci-' cent., said to be obtained, shows a great economy in working. These figures, furnished me by the proprietor of the forges, Mr. Putnam, were afterwards confirmed by Mr. Pearson, the director of those at East Miil- dlebury, where the very rich ores from the same region are treated. The dimensions and construction of the New Russia forges have already been given. The pressure of blast employed was from Ih to It pounds, and the average produce of iron for each fire, 2,400 pounds of bloom-iroi> in twenty-four hours; the amount of charcoal consumed being from 2o(i to 000 bushels to the ton of blooms produced, and the weight of the charroal from sixteen to eighteen pounds to the bushel. At East Middlebury, where, as just stated, the conditions are very similar, the estimated consumption of charcoal was 270 bushels to the ton of blooms, a result which is the mean of the figures obtained at the New Russia forges. Some of the ores here used contain a little phosphate of lime, and it was observed that when too hot a blast was used, although tlic production of metal was rapid, the iron from these ores was hot-slinri. while 'vith the cold blast, formerly employed, the iron, although produce,! ir;>re slowly, was never hot-short. The force of the blast at these forges was equal to one and three-ijuarter pounds, and even two pounds to the inch. Mr. Pearson, the director of the East Middlebury forges, made, in the autumn of 1807, experiments on several tons of the iron sands from Seven Islands, page 200, and succeeded in obtaining from them about three- eighths of their weight of good iron. lie, however, found it necessary, in order to treat these fine sands, to reduce very much the force of the blast, an experience which has been confirmed by the practice at Moisie. It Waste beat. Now Kussi» I'orgc-. Kii^l Mlddl.- bury lorge.'. MoUie forges. Consiirapli" 01" cluire-Ml. Blief> of heartUa. 280 GEOLOGICAL SURVEY OF CANADA. [36 appears to be from ignorance of this fact, that the bloomers of New York had always rejected the fine sandy ore separated during the process of Avashing, as being unsuited for treatment in the bloomary fire. At Moisie, although eight forges have been constructed, but four of them were in operation at the time of my visit in August, 1868, and the same number, I am informed, in October last, two of the furnaces not having yet been completed. A reverberatory furnace has, since my visit, been constructed, in which it is proposed to re-hcat the loups for the second ham- iiioiing, instead of returning them, as in most cases is done, to the forge-fire for that purpose. The opening of the twyers used measured one inch by one and seven-eighths ; they were inclined downwards at a very small angle, it having been found by experience that the considerable inclination which is used with the coarser ores cannot be advantageously employed with the fine sands. In like manner, as remarked above, it has been necessary to reduce the force of the blast, to from f to li pounds, the iverage working-pressure being about one pound to the inch. According to the latest accounts, there were, in October, four hearths in regular opera- tion, requiring four bloomers, one assistant to furnish coal, etc., and one liaramerer, being six men in all for each shift of twelve hours. Each hearth furnished eight loups daily, and the aggregate yield of iron was estimated at three tons, or three-quarters of a ton for each hearth, every twenty-four hours. The consumption of charcoal was 1400 bushels daily, being at the rate of 460 bushels to the ton of blooms, or 350 bushels to each fire. This charcoal is chiefly produced from spruce and fir, with some admixture of birch, the wood being mostly small, and the weight of the coal is stated to be fifteen pounds to the bushel. This gives a consumption of G990 pounds of charcoal for the production of 2240 pounds of blooms, being at the rate of 3.12 pounds of charcoal for the pound of iron. If we compare this result with the figures given above, for those forges wbich treat nearly pure magnetic iron ores, we find that to produce a ton of blooms there are consumed, at East Middlebury, 270 bushels, and at New Russia from 250 to 300 bushels of charcoal, weighing from sixteen to eighteen pounds to the bushel. If we assume, in both cases, the greater weight, of eighteen pounds to the bushel, we have for 250 bushels, 4500 poimds, and for 300 bushels, 5400 pounds of charcoal, the former corres- ponding to 2.01 pounds, and the latter to 2.41 pounds of charcoal to the pound of iron, or, taking the mean of the two, 2,21 pounds, as compared with the 3.12 pounds said to be consumed at the Moisie works. If now, we consider the relative sizes of the difi'oreut bloomarv hearths, we find them to be as follows : — New Russia 20 x 32 iuches = C,400 square ioches. E(i9t Middlebury 24 X 29 ,, = 0,900 ,, „ Moisie 30 X 32 „ = 9,0U0 „ „ [36 f New York 3 process of four of them ind the same 3 not having Y visit, been second ham- the forge-fire I one incli by a very small Ic inclination ily employed it has been : pounds, the According 3gular opera- etc, and one lours. Each I of iron was loarth, every )U3hcls daily, >0 bushels to with some weight of the consumption s of blooms, of iron. If lose forges uce a ton of and at New sixteen to the greater shels, 4500 mcr corrcs- ooal to the ,3 compared .rv hearths. hies. 37] REPORT or DR. T. STF.RRY HUNT. 281 The area of the Moisie hearths 's, then, in round numbers, one and a-half times tha*; of the others, and, with an equally powerful blast, they should consume one-half more charcoal. This increased size is, however, counter- balanced by the feebler blast, and we find that each fire at Moisie con- sumes, in twenty-four hours, 350 bushels of charcoal, equal to 5250 pounds, which, from the calculations already given for the New Russia forges, should produce, with an ore such as there treated, 2375 pounds of iron. In fact, the Moisie forges, according to the data before us, with an area one-half greater, consume daily the same weight of charcoal as those of New Russia, and produce only two-thirds as much iron. I have very recently been informed that, with careful management, it has lately been found possible so far to reduce the consumption of fuel at Moisie, that a ton of blooms can be made with 350 bushels of properly prepared charcoal. The consumption of ore, which formerly amounted to three tons or more for a ton of blooms, is also said to have been considerably reduced, the daily production of iron from each hearth, however, remaining the same as before. The cause of this small production of iron, as compared with the area causes oniio of the furnace, and with the consumption of fuel, is not, in my opinion, to '■"'»">-''• y"-'''- be found either in the reduced force of the blast or in the mechanical condition of the ore. A great heat is not required for the reduction of the o.xydof iron to the metallic state, and other things be equal, the finer its subdivision, provided it be not dissipated by the blast, the more rapid and more complete should be its conversion to the condition of metal, by the action of the reducing gases, as it passes downward through the mass of burning charcoal. Such coarse grains of ore as pass, incompletely reduced, through the ignited fuel, and in this state reach the slag below, have no chance of further reduction in the forge. Ilcnce we may conclude that, the fineness of the ore, should, under favorable conditions, render the reduction more complete. The principal cause of the small yield of the Moisie furnaces is appa- rently to be found in the incompletely purified condition of the ore. It will Nature or orci be seen in the detailed analyses on page 267, that the iron sand, as now prepared for the forge, may, by the use of the magnet, be divided into two nearly equal portions. One of these is magnetic, and consists, for tho greater part, of magnetic oxyd ; it contains over two-thirds its weight of iron, and is nearly equal in richness to the magnetic ore used in the New Russia forges. The other half is a highly titaniferous oxyd, mixed with more or less silicious matter, and containing only 44 per cent of iron ; and its admi-xture with the magnetic oxyd, which reduces the proportion of iron in the whole to 55 per cent, appears to be not merely useless, but actually prejudicial. 282 GEOLOGICAL SURVEY OF CANADA. [38 giliclous inipii rltiw. When an impure ore of iron is treated in the blast-furnace, certain substances, called fluxes, are added, which form fusible combinations with impurities. Thus, if the ore contains P'^'''a, a sufficient quantity of lime is smelted with it, and a silicate of lime is formed, while the oxyd of iron, being left free, is wholly reduced to the metallic state. In the direct method, on the contrary, no fluxes arc used, and if silica be present in the ore, it combines with a portion of the oxyd of iron, forming a silicate of >aturc of slug.". iroH, which melts into a slag or cinder, from which the iron cannot be sepa- rated in the forge. Thirty parts of silica will, in this way, unite with soventy-two parts of protoxyd of iron, equal to fifty-six parts of metallic iron. In the case of the somewhat silicious ores of the Pyrenees, treated ill the Catalan forge, we have seen that three-tenths of the iron present ill the ore pass into the slag, and the loss would be much greater did not these ores hold a considerable proportion of manganese, lime and other bases, which help to satisfy the affinity of the silica, and to leave the iron tree. Such substances as these, play the part of fluxes with a silicious ore, liut if they are wanting, a portion of the oxyd of iron itself is consumed lor the purpose, forming, in fact, the only flux for the silicious impu- rities, when such an ore is treated by the direct method in the bloomary fire. Whenever, in the Catalan forge, the American bloomary fire, or any other direct method, we have to treat an ore containing free silica, provided other bases are not present, we must always allow oxyd of iron, in the proportion already indicated, for the saturation of the silica, being at the rate of nearly two parts of metallic iron for each part of silica present in the ore. It is for this reason, it may bo remarked, that kiln-burned charcoal is to be preferred, for the bloomary hearth, to charcoal made in piles ; the latter being generally more or less impure from adhering silicious earth, which, by combining with o.xyd of iron, causes a waste of the ore. The quartzose sand which is mixed with the iron sands, is nearly pure silica, and the oxyd of titanium which they contain, appears, from the analyses of slags given below, to require, for fluxing it, as much oxyd of iron as the silica itself. These slags, in case no other bases than oxyd of iron are present, should approach very closely to. the composition of a tribasic silicate of protoxyd of iron, which, as already explained, contains 30 of silica to 72 of proto.xyd of iron, or 29.40 per cent, of silica, and 70.00 of protoxyd, equal to 54.1) per cent, of metallic iron. The highly titani- ferous slags produced at the Moisie furnaces, contain, in some oases, a still large proportion of oxyd of iron. Of the following analyses, I is of a crystalline, black, brilliant magnetic slag, which contained canties lined with large pyramidal crystals, apparently dimetric in form. It was produced at tlie Moisie forges in the autumn of of 1807. II was a portion of the ordinary slag produced at the time of [38 CO, certain ations with ity of lime :yd of iron, the direct (Sent in the L siUcate of lot be sepa- , unite with of metalUc jes, treated ron present iter did not ! and other ,ve the iron iilicious ore, 13 consumed icious impu- le bloomary lary fire, or free siUca^ Kyd of iron, ca, being at ica present dhi-burned )al made in ing siHcious of the ore. icarly pure from the )xyd of iron .vyd of iron f a tribasic tains 30 of and 70.G0 hly titani- le cases, a magnetic ipparontly I autumn of Ihe time of 39] REPORT OE DR. T. STERKY HUNT. 283 vcij visit, in August, 1868, and was similar to the last, but somewhat moimc vesicular, the cavities being lined with very small brilliant crystals. Both of these slags readily gelatinized when treated; in powder, with hydro- chloric acid. The residual silica, however, showed a portion of grains of undecomposed ore, which was larger in the second specimen ; it was, in each case, deducted from the analysis. The whole of iron in both of these slags is represented as protoxyd, and the results are compared with those of two analyses of the non-magnetic portion of the ore, copied from pages 267 and 268, and here given under III and IV. Protoxyd of iron . . Oxyd of manganes'" Lime Magnesia Alutniua Titanic acid Silica I. II. III. IV. G7.14 52.31 53.20 50.38 undet. 2.04 1.10 1.37 .95 .80 .18 . . • .56 .... • • • 20.07 34.05 30.74 28.95 8.75 11.29 6.14 8.75 Metallic iron. 98.13 52.22 100.42 40.68 45.20 43.85 From a comparison of the above analyses it will be seen that the first slag contains more oxyd of iron than the non-magnetic portion of the ore ; which? in the conditions of working, at the time the slair was produced, actually dissolved and carried away a considerable portion of the reducible ore- If we were to regard one half of the washed ore as composed of pure magnetic oxyd, this, were it wholly reduced, could only yield an amount of metallic iron equal to 36 per cent ; but the magnetic ore, as we have seen, still retains more than 6 per cent of silica and titanic acid, which must 1)0 removed by fluxing with a portion of the o.xyd of iron present, givhig rise to a certain amount of slag. Meanwhile the non-magnetic ore, in molting, removed another portion of iron oxyd, so that when this slag was made, more than three tons of a mixed ore, having the composition aV)ovo given, must have been consumed for the production of a ton of blooms ; while, of the magnetic portion of the ore, one and a-half tons, or a very little more, would suffice. (In the production of the slag II the loss of iron was somewhat less.) This explains why the Moisie furnaces have yielded, when compared with those of New York and Vermont, so small an amount of iron for the labor employed and the fuel consumed. To produce a ton of iron it has been necessary to handle twice as much ore as i.s required in forges where a pure ore is treated, and moreover one and a-half tons, or more, of worthless material have been fused, and got rid of as slag, thus involving a great waste of fuel, as well as of labor. It may here be remarked that a portion of slag taken by me from the East Middle. 284 GEOLOGICAL SURVEY OF CANADA. [40 Reduction in crucibles. bury forges, contained according, to Mr. Broome's analysis, 48.2 per cent of iron (equal to 62.06 of protoxyd), and 16.70 of silica, besides 17.33 of alumina, and 1.82 of oxyd of manganese. The amount of slag produced by the rich ores which are treated at these forgeS,is comparatively very small. It would seem probable that by a judicious management of the working, the waste of iron in the slags at Moisie, might be considerably reduced, and this result, we are assured, has lately been attained ; but it will still remain true, that a large amount of iron-oxyd must be consumed to flux the considerable proportions of silica and titanic acid, which are present in the mixed ore, even after careful washing. It should here be explained that the result would be far otherwise if this ore, with all its impurities, were to be fused in a crucible with carbon- aceous matters, with, or even without proper fluxes. In the former case, as in a blast-furnace, the whole of the iron which it contains, amounting to not less than 55 per cent., might, by judicious admixture, be set free, and reduced ; and in the latter cases, without fluxes, it has been shown by Percy, that by fusion at a high temperature, in a crucible lined with charcoal, the tribasic silicate of iron, already noticed, gives up two-thirds of its iron, which is reduced to the metallic state, so that the amount of unreduced oxyd retained by the slag would be inconsiderable. From this it is evident that the results of fire-assays, or trials on a small scale in crucibles, cannot serve as a guide to the working of iron ores in the direct method. A certain amount of lime added to the ore, would doubtless reduce the waste of iron in the slags, and thus allow more iron to be obtained from the mixed ore ; but although such an addition is useful in the blast-furnace, it would require experiments to determine whether the practice could be advantageously introduced in working in the bloomary-hearth. In a region where the ore is so abundant and so cheap as it is at Moisie, the saving of iron is a consideration which should be subordinate to the economy of fuel and labor, and the most profitable way of working these iron-sands would seem to be by separating and rejecting the non-magnetic portion, by some apparatus like that described farther on. Quality ofironj The quality of the iron produced at the Moisie forges is superior. As the result of experiments made upon it in England, it is said to possess a tensile strength greater than that of Low Moor iron, and to work easily both hot and cold. It is now employed at Montreal for the manufacture of railway axles. The fact that those objectionable elements, sulphur and phosphorus, occur in but very small quantities in the iron-sand of Moisie, has already been noticed. It is probably *o the absence of these that the excellence of the Moisie iron is due. In a specimen taken from a bloom which was [40 i.2 per cent ie8 17.33 of )roduced by ' very small, he working, ly reduced, t it will still d to flux the eaent in the Twise if this rith carbon- •ormer case, , amounting be set free, }n shown by I lined with p two-thirds 1 amount of From this lall scale in n the direct reduce the tained from ast-furnace, e could be In a region the saving conomy of iron-sands |tic portion, |)enor. As possess a york easily [anufacture [hosphorus, las already lexcellence Iwhicb was 41] REPORT or DR. T. STERRY HUNT. 285 ary iron. made in my presence, at the Moisie forges, the presence of sulphur could be detected by delicate tests, but its amount was only .0094, or less than To.ios > while the quantity of phosphorus present was equal to '0184 per cent. This iron contained no trace of titanium in its composition, and a small mass of white crystalline cast iron, which had accidentably been formed in one of the forges, was equally destitute of titanium. The cost of producing a ton of iron blooms directly from the ore, by the cost of^bioon»- bloomary process, varies greatly with the price of the dressed ore, which will depend on the proximity of the mine to the forge, and the richness of the crude ore. Thus, the cost of the two tons of dressed ore employed to make the fine iron of the Ausable forges, was estimated by Mr. Rogers, in 1868, at not less than |>18-00, while the one and a-half tons of ore con- sumed at New Russia, would not probably cost more than one-half that sum. The followmg estimate made by a highly competent iron-master, in 1868, may serve as a guide to the cost of producing iron at that time in New York : — 2 toni of ore $10.00 300 bushels of charcoal (& 8c 24. 00 Wages 9.00 Gtneral expenses 3.50 Cost of the ton of blooms $46.50 The above prices are in American currency, which, at that time, was equal to about j^^, making the gold-value 837-20. The estimate of another manufacturer, in Clinton county, gave -f 7*00 for wages. It will be observed, moreover, that the amount of charcoal, in the above estimate, exceeds the average consumption for the production of a ton of blooms, which may be taken at about 270 bushels. To produce a ton of blooms from cast iron, in what is known in Sweden, as the Lancashire hearth, there are consumed, according to an authority cited by Percy, 28 cwt. of pig iron, and j\ tons of charcoal. In New Jersey and Pennsylvania the conversion of the pig iron, is, for some purposes, efiected by a somewhat similar process, which involves two operations, the melting in the running-out fire, and a subsequent treatment in the sinking-fire, as it is called, which is a bloomary forge very like that used for the ore in the direct method. To produce a ton of blooms in this way, there are consumed 24 cwt. of pig iron, and 100 bushels of charcoal, according to one authority, while another estimate gives 120 bushels ; the quantity varying both with the quality of the crude metal, and the charcoal ; while, with some arrangements, the consumption of fuel is much Comparative cost. 286 GEOLOGICAL SURVEY OF CANADA. [42 I'urifying ore^. gi'eator. The mean of these, 110 bushels, at 18 pounds to the bushel, would give, almost exactly j\ of a ton, the amount used in Sweden. The (juantity of charcoal consumed for the production of a ton of pig ii'on in the United States varies greatly, but in the best constructed and more modcra furnaces, like those of Michigan, with rich ores, will not exceed 130 bushels of charcoal of the above weight, which gives, for 24 cwt. of pig iron, 156 bushels. (See page 256.) This, added to 110, equals 266 bushels, the total amount of fuel required to produce a ton of blooms by means of the blast-furnace with the charcoal-finery. There would appoar to be but little difference, so far as the consumption of the fuel is concerned, between the cost of producing bloom-iron by the direct and indirect methods just described. The first cost of the establishment for the former is, however, less, and this is probably one of the reasons which has led to the adoption of the direct method by the bloomary forge in northern New York. The conversion of the oxyd of iron to the metallic state, under t^ie influence of solid carbonaceous matter, or reducing gases, takes place at a temperature considerably below that at which the affinity of silica for the oxyd of iron is exerted. Even the compound of titanic acid with oxyd of iron is decomposed at a red heat in contact with hydrogen gas, the iron being wholly reduced to the metallic state. If it were possible to cflfect this reduction, and subsequently to eliminate the silica and titanic acid from the metallic iron, ores containing these impurities might be made available for the direct method of conversion ; but the practical difficulties of effecting such a separation are such that the only available modes of treating such ores as contain considerable amounts of these impurities, are to smelt them in the blast-furnace with proper fluxes, or to eSect as complete a separation of the impurities as possible, before submitting them to the process of reduction. This, in the case where heavy granular ores are mixed with quartz and feldspar, as for example, at the Palmer ore-bed, already noticed, is attained by washing away the lighter materials. Where, however, the impurity is chiefly titaniferous iron, as in the Moisie sands, the separation may be readily efiected by means of magnets, a process which is equally advantageous where magnetic iron ore is mixed with lighter impurities, as quartz or silic'ious minerals. The use of magnets for this purpose has long been taken advantage of, and various machines with permanent and with electro-magnets have been contrived. A simple and ingenious arrangement for this end, which has been invented and patented by Dr. F. A. H. Larue, of Laval University, Quebec, appears to be novel in the mode of its working, and is very efficient and cheap. The mixed sand or crushed ore is poured through a screen, into a hopper, the discharge of which is so arranged as to open and close at proper intervals of time, and, falling from this, is spread [42 :ho bushel, n Sweden, of pig iron [ and more not exceed 24 cwt. of 166 bushels, y means of ,r to be but (d, between ethods just s, however, le adoption 'ork. , under the 1 place at a lica for the with oxyd in gas, the possible to and titanic it be made diflJculties modes of mpurities, effect as Itting them nular ores r ore-bed, Where, sie sands, a process ixed with ^ntage of, ^ave been rhich has diversity , is very irough a td as to ps spread 43] REPORT OF DR. T. STEREY HUNT. 287 111 a thin and uniform layer, upon a series of aprons arranged, with Lame'!' nmp- 1 , 11 1 11 1 i 1 ■ 1 '1''*''^ madiiue. interspaces, between two parallel endless bands, which pass over two horizontal cylinders. These aprons, charged with ore, are made, by the movement imparted to one of the cylinders, to pass from beneath the hopper, and under a series of permanent horse-shoe magnets, 800 in number, each capable of sustaining about five pounds weight, arranged upon transverse bars, in five rows of IGO magnets each. Beneath these is a tympan, covered with muslin, which, when the iron ore is passing beneath tliem, is in the contact with the poles of the magnets. So soon, however, as the magnetic portions of the ore have arranged themselves, by magnetic attraction, in adhesion to the under side of the tympan, and the apron has moved from beneath, and gone forward to discharge the non-magnetic portion of the ore at the foot of the machine, the tympan is momentarily withdrawn a short distance from the poles, and the adhering magnetic ore falls in the open space between two aprons, into a receptacle placed below. This process of loading and unloading the magnets can be repeated twice in each minute. These machines, as now constructed, occupy a space of about six feet by five, and are four feet high ; they are said to cost, at Quebec, at about 8-300 each. One, of these dimensions, will, according to Dr. Larue, treat in an hour, three tons of sand holding one-third of magnetic ore, separating from it one ton, containing over ninety-nine per cent of magnetic grains. I have myself seen only a smaller machine, the first one constructed, which had a capacity of about onc-lfalf that just stated. The motive [inwer required is very small, and the mechanism, as will be seen from the description, exceedingly simple. Dr. Larue observes, that, inasmuch as a rich sand may be passed through the machine as rapidly as a poor one, the yield is directly proportionate to the amount of magnetite present, so that a sand containing one-fourth as much as that above mentioned, would yield about six tons of purified sand in twenty-four hours. Even very l^uor sands may, probably, with this machine, be treated with advantage. The same process of purification may doubtless be applied with advantage, after crushing, to the preparation of lean massive magnetic ores for the bloomary fire, or for other direct methods for conversion into iron and steel. A process of partial reduction, at a low red heat, will render non- magnetic iron ores attractable by the magnet, a reaction of which Clienot long since proposed to take advantage, for the purification of such iron ores as are not naturally magnetic. In accordance with the well-known fact that the reduction of oxyd of iron takes place at a temperature very much below that required for its subsequent carburation and fusion, it has been shown that the charge of ore in a blast-furnace is converted to the metallic state some time before it 288 GEOLOGICAL SURVEY OF CANADA. [44 Chcnot'8 zuethod. JroMMw.'"** descends to the zone iu which melting takes place. It forms, when reduced, a spongy mass, readily oxydized, which, by proper mana^^ement, can be compressed and made to yield malleable iron, or by appropriate modes of treatment, may be converted into steel. This fact has been the starting point of a great number of plans designed to obtain malleable iron and steel, without the production of cast-iron and the employment of the processes of puddling and cementation. This, it is true, is attained in Catalan and bloomary forges, but tcntion of many inventors has been, and still is, directed to the i ^very of simpler, or at least of more economical methods of obtaining similar results. A short sketch of the various new processes will not be without value, as bearing upon the utilization of the iron ores of Canada, and especially of its iron sands. Of these, the method of Chenot is best known. His expfriments seem to have been commenced about forty years ago, since we are informed that he had erected a large furnace for the direct treatment of the ores of iron, in 1831, although his results were not brought before the public until twenty years later, at the International Exhibitions of 1851 and 1855. I was a member of the International Jury at the latter, and had an oppor- tunity of studying Chenot's process as then conducted, on an industrial scale, at Clichy, near Paris. A description by me of the process as then and there practised, will be found in t'"> report of the Geological Survey for 1855-57 (page 397). Rich peroy "es were broken in small pieces, mixed with a portion of charcoal, !> ced in large vertical rectan- gular muffles or retorts, enclosed ia a gas-furnace, and heated to redness. The ore, after being reduced to the state of metallic sponge, passed down- wards into an air-tight cooling-chamber, which was a continuation of the muffle, and when sufficiently cooled, was withdrawn. The spongy metal, thus obtained, was then exposed to a welding heat in a proper furnace, and formed into balls, which were afterwards treated like the balls from a puddling-furnace, and gave malleable iron. By impregnating the metallic sponge with oily and tarry matters, and afterwards expelling these by heat, a sufficient amount of carbon was fixed in the metallic sponge to convert it into steel. By grinding, compressing and melting this carbon- ized sponge, cast-steel of a superior quality was manufactured at prices which, it was claimed, were much below the cost of steel prepared by cementation of bar iron. This process was subsequently introduced ia several places in France, Belgium and Spain, where it was applied to the manufacture of bar iron, and up to 1863 at least, was worked on a con- siderable scale at Baracaldo, in Spain, where, in 1859, about ten tons of iron were manufactured daily from iron sponge. A very important modification of the process already described, in which the heating was effected externally and indirectly, consisted in the [44 1 reduced, nt, can be e modes of ho starting e iron and ent of the ittaincd in 'enters has at least of t sketch of ig upon the sands, ments seem e informed the ores of aubUc until A 18.35. I i an oppor- 1 industrial less as tlieu ical Survey paall pieces, cal rectan- to redness. 3sed duwn- tion of the ngy metal, rnace, and lis from a ,e metallic these by sponge to lis carbon- at prices ipared by iduccd in ,ied to the a con- in tons of [ribed, ia led in the 45] REPORT OF DR. T. STERRY HUNT. 289 Cimiparativei cost. internal or direct method of heating. In this the outer furnace and the chenot'g direct admixture of charcoal with the ore were both dispensed with. The vertical '^" "" reduction-chamber wa filled with ore only, which was reduced by the action of currents of heated carbonic oxyd gas, obtained by forcing air, at a pressure eJjual to half an inch of mercury, through two gene- rators filled with ignited charcoal. This mode of producing the sponge was found much more economical than that by indirect or external heating. The working results of the direct method, as carried on at Lamarade, in Spain, in 1863, are given by Percy ; from which it appears that for the production of one ton of blooms, there were consumed 1.87 tons of charcoal. The greater part of the fine Swedish iron used at Shef- field for the manufacture of stee', is produced from charcoal-made pig, treated in a charcoal-finery, known as the Lancashire hearth, and is obtained with a consumption of charcoal, which, for the united processes of reduction and refining, amounts to 1.90 tons for the ton of blooms, a result almost identical with that of the process of Chenot. (Percy, 3Ietctl- ^"''^^) PP- 342-596.) The modified Catalan forge, and the American Moomary fire, as we have seen, produce malleable iron with a consumption of charcoal which is not very much greater, and with a simpler, and probably less expensive apparatus than that required for the Chenot process ; while the method by the blast-furnace permits of the use of ores which are unfit for trcatmt'it by any of these direct processes. Tlie pat< ^s granted to Clay, in England, in 1837 and 1840, were for the manufat '•e of mailable iron by a process essentially the same with Chcnot's eari. method of indirect or external heating. According to Clay, hematite ores were mixed with one-fifth of their weight of charcoal, coke, or other carbonaceous matter, and heated to bright redness in a clay retort, or other suitable vessel, until the ore was converted to the metallic state. When the reduction was complete, the spongy iron (without previous cooling, as in Chenot's plan,) was transferred directly to a puddling- furnace, where it was brought at once to a welding heat, made into balls, and then wrought into blooms in the usual manner. This process was tried on a pretty large scale near Liverpool, in 1845-46, and altliough iron was regularly made by it for some time, and to the amount of 1000 tons, the process was not found to be commercially profitable, and was abandoned. The process of Ronton, patented in the United States in 1851, was Ronton'e pro- very similar in principle and mode of working to that of Clay. The mix- '^''■*' turc of ore and coal was introduced into a vertical mufile or retort, which was inclosed ^in the flue or chimney of a furnace, not unlike an ordinary puddling-furnace. The contents of the muflie, being suflSciently heated, were reduced to the metallic state, and, from time to time, discharged from Clay'i metlioU 290 GEOLOGICAL SURVEY OF CANADA. [46 IIarvpy'8 pro- cess. HH! Guilts patent, the bottom, into the furnace, where the spongy iron was exposed to a welding heat, and wrought into blooms. This process, after having been essayed on an industrial scale at Cincinnatti, and at Newark in New Jersey, was abandoned. A similar fate attended the trials, on a large scale, of Harvey's patented process, at Mott Haven, near New York, about the same time. In this, the coarsely powdered ore, mixed with charcoal, was placed on inclined trays or shelves of stea*-itc, in a heated chamber con- nected with a welding or balling-furnace. The flame from a fire below was made to pass through the chamber, and the ore, being at length reduced to the metaUic state, was transferred to the hearth below, and there converted into blooms. For a farther description of these various pro- cesses, and the similar plan of Yates, the reader is referred to Percy's Metalhir'i'/, pp. 330-348. Chenot's plan of reducing the ore by a current of carbonic-oxyd gas, was adopted by Gurlt, who used the direct mode of heating, already noticed. The gases from the generators charged with fuel, were led through flues, into the vertical reducing-chamber, a blast of air being at at the same time introduced into the flues, in sufiicient quantities to keep up the combustion of the gases. By this means, according to the speci- fication, " there passes into the shaft a mixture of flame and carbonizing and reducing gases, by which the iron ore is heated" and carbonized. According to Gurlt's patent-specification, (No. 1G79, London, July I'i, 18oG,) by continuing, for a sufficiently long time, the action of the gases, the resulting iron sponge may be more or less carbonized, so as to yield, by subsequent fusion, either cast iron or steel. These partially carbonized products he proposed to melt in a reverberatory gas-furnace, the blast of air into which is to be " so regulated that it exactly burns the gas produced in the generators," and that neither unburned gases nor unconsumed air escape ; the object being to obtain a neutral flame, which should not alter the sponge upon the hearth. In this way carbonized sponges from rich ores, are said to have been successfully converted into cast iron iii Spain. Gurlt's ingenious specification thus involves the idea of first reducing the iron ore to a metallic sponge, and afterwards carbonizing this sponge, so that, by subsequent fusion, it may be converted into cast iron or steel. Although the conception of thus carbonizing the iron while in a spongy state, is probably novel, the use of carbf^naceous gases or vapors fur carbonizing iron, and converting it into steel, is not new, as may be seen from the patent for this purpose granted to Macintosh in 1825. The experiments of Percy upon iron wire have also shewn the rapid carbonizing effect of coal-gas and heavy oily vapors, like those of parafline ; (^Metallur<jy, pages 109 and 773) and, according to Marguerite, carbonic-oxyd gas, at [46 osed to a ving been I in New u'ge scale, about the ircoal, was imber con- fire below th reduccil and there arious pro- to Percy's ;-oxyd gas, ig, already , were led ir bemg at ies to keep i the speci- carbonizing carbonized. 1, July 1(3, the gases, yield, by carbonized le blast of \ produced sumed air hould not liiges from bt iron in reducing |s sponge, or steel. a spongy ipors for be seen 5. The •boniziug allunjy, 11 gas, at 47] REPORT OF DR. T. STERRY HUNT. 201 an elevated temperature, yields up a portion of its carbon to iron, which is thus converted into steel. Practical diflSculties have hitherto prevented the application of hydro-carbon gases and vapors to the carbonizing of bar iron on a large scale. With the results of Chenot, Gurlt, and Macintosh before us, we are prepared to understand the process of Dr. George Hand Smith, of Rochester, New York, which is just now attracting some attention in the United States, for the production of steel. The crushed and purified ore, or iron sand, mixed with a portion of pulverized charcoal, is heated in a kind of reverberatory furnace, with an arrangement which permits the vapor of petroleum or coal-tar to pass through the mass, thus aiding in the reduction, and finally carbonizing the resulting sponge, which is then trans- ferred to a puddling-furnace, to be wrought into iron, or, if properly car- bonized, into steel. Before proceeding farther, mention should be made of some other nans for work- iiig iron sauds. methods which have been devised for the treatment of iron sands, and for their conversion into iron or steel. In 1851 a patent was granted to Stenson, for a process for working the iron sands of New Zealand, and similar ores from India. These were to be mixed with small portions o[" clay and lime, with or without the addition of charcoal ; the mixture wa^ ground in a pug-mill, with water, and formed into lumps, for subsequent treatment in the blast-furnace. In 1862, Moreau proposed to mix ir.ai sands with iron filings or turnings, and then incorporate them with fuel, such as peat-coal or coke; the mixture being made into blocks, which wore to be smelted in suitable furnaces. In 1866, Mr. James Hodges, avIio was not acquainted with the experiments of Moreau, moulded the iron sands of Moisie into blocks with peat, and by treating these, after drying, in a proper furnace, succeeded in converting the oi'e into malleable iron, at a single operation. (^Report of G^olo(jical Siirvi'i) for 1866, p. 201.) Messrs. Whelpley and Storer of Boston eifect the reduction of the iron sand ore, or pulverized ores, on the hearth of a reverberatory fuinace, which is heated, in part, by pulverized coal, borne by a blast of air over the fire of solid coal upon the grate. In this way the furnace-cham'ior is filled with a volume of burning coal-dust, which can, by regulating the supply of coal and of air, be made either oxydizing or reducing. The heated ore upon the furnace-hearth is thus reduced to the metallic state, balled and made into blooms, with, it is claimed, a great economy of fuel. It has also lately been proposed to convert these sands into steel or cast iron, by melting with a sufficient admixture of charcoal in crucibles, or other closed vessels, heated from without. This is, in fact, nothing moro than an extension of the dry method for assay of iron ores. A patent for making steel in this way, by treating rich ores, mixed with carbonaceous 292 GEOLOGICAL SURVKY OF CANADA. [48 Ponsard'a results. Steel direct from matter, in air-tight melting-pots, was granted to Lucas, in 1791, and a similar claim was uiade by David Mushet, in 1800 ; while, according to Percy, " experiments in the direct production of cast steel from iron ores, in crucibles, were made by Riley, at Dowlais, a few years since, and although excellent steel was occasionally produced, it was not found pos- sible to ensure uniform results." (^Metallurgy, p. 765.) More recently, Ponsard has brought forward a similar process, the results of which were communicated to the French Academy of Scieiir^s, July 19, 1809. This arrangement consisted of a number of fire-proof crucibles, about eight inches in diameter and forty inches high, which were placed in a reverbcratory gas-furnace, the mouths of the crucibles being fitted into openings in the furnace-roof, for convenience of charging. The lower part of the crucible is perforated, and rests on the sole of the furnace, which is furnished with gutters leading to a depression or basin in the middle of the furnace-hearth. The crucibles are charged with the ores, mixed with proper fluxes, and about twelve per cent, of carbon, sufli- cient to effect the reduction and carburation of the iron, which, under the influence of a very intense heat, melts, and, running through the holes in the bottom of the crucible, collects in the basin in the middle of the fur- nace. According to Ponsard, a ton of coal is consumed for each ton of iron produced, so that the process cannot be recommended for its economy of fuel. He, however, claims as a great merit of this process, the complete separation of the fuel from the carbon required for the reduction of the ore, so that for the furnace, inferior kinds of combustibles, which, if brought directly in contact with the ore, would injure the quality of the metal, may be used with safety and advantage. The process patented by Johnson, Jan. 22, 1808, as described in the Practical Mechanics' Journal for June, 1809, (quoted by Osborn in his Metallurgy of Iron and Steel, page 808) is, however, exactly similar, in all its details, to that of Ponsard, which was first announced as a novelty to the French Academy, July 19, 1809, eighteen months later. In a spccifica- ijirue's patent, tion dated at Quebec, July 10, 1809, Dr. Larue claimed, and subsequently received letters-patent for Canada, for a process similar in design to that of Johnson, of which he was ignorant. Although there wore differences in detail, the avowed object in both plans was to separate the ore, with the carbon required for its reduction, from the fuel, (which might, conse- quently, be of tui inferior quality,) and to permit of a continuous charging and discharging of the crucible. The difficulty of constructing sufficiently refractory crucibles for the intense temperature, and the small yield to be expected from such a process, would perhaps prevent it from over being used for the manufacture of cast iron. Dr. Larue, hovrever, anticipated its api)lication to the production, not of cast-iron, but of cast-steel, which Johnson's Iiatent. [48 49] REPORT OF DR. T. STERRY HUNT. 298 91, and a ording to iron ores, since, and found pos- ■ocess, the ■ ScieurflS, ; fire-proof vhich were ibles being ging. The sole of the •n or basin cd with the irbon, suffi- , under the lie holes in of the fur- each ton of its economy he complete stion of the , if brought the metal, [bed in the [l}orn in his similar, in novelty to la specifica- Ibsequently Vign to that llifferences !, with the [ht, conse- charging lufficiently }iold to be Iver bciug Inticipatcd Bcl, which would require a very nice adjustment of the proportions of carbon to secure a uniform quality in the product ; as in the ancient processes of Lucas and Mushet, and the more recent experiments of Riley, mentioned by Percy, and referred to above. Two processes for the production of steel are those which depend, caststoeu. respectively, on the combination of cast iron in proper proportions with malleable iron or iron sponge, and with oxyds of iron. In the specification of a patent granted in 1839, Heath claimed the production of steel by Heatu'spat«nt«. melting with cast-iron, 3ither wrought iron, or oxyds of iron or manganese. In a second patent, granted to h xi in 1845, he described an arrangement by which the cast-iron was kept in a molten condition, in a gas-furnace, while pure iron in scraps, or in sponge, obtained by reducing oxyd of iron, as in Chenot's and Clay's method, was added from time to time ; until, by trial, the proper quality of metal had been obtained, after which the liquid steel was run into ingots. Other processes, based on the reactions embodied in Heath's first patent, are those of Uchatius, (patented in 1855,) who melts granulated cast iron in crucibles, with a certain proportion of pure oxyd of iron, and thus obtains a fine quality of steel, (a process already specified in Wood's patent, in 1701) ; and that of Brown, (patented in 1856) who, to produce steel, melts, in crucibles, mixtures of pig iron and clipped bar iron. This method is practised to some extent in Sweden, where it is known as the Obersteiner process. In the process of Obuchow, which appears to be successfully used in obuchow-a Russia, fine pig-iron is melted, and run into a large crucible, previously heated to whiteness, and holding magnetic iron ore, alone, or with titanic iron sand and iron and steel scraps. The crucible is then heated till the contents are perfectly fluid, some nitre and arscnious acid are added, and the steel run into ingots. By a somewhat similar process to this, Ellers- hausen attempted to produce steel, by pouring molten cast-iron upon previously oxydized shcet-irou, heated to redness, and placed in a heated Kiiorehawcun vessel. The oxyd dissolved in the molten iron with violent chemical '' '" action, decarbonizing it, and producing a kind of steel ; but it would pro- bably bo difficult to effect a thorough conversion of the iron without keeping up the heat from without ; which was not done in Mr. Ellershausen's first experiments, made in Montreal, in the spring of 1868. The above processes, however, involve the use of crucibles, and it had become a great desideratum to produce cast steel upon the open liearth. This was the aim of Heath, in his process described above ; but the difficulties in producing and controlling a heat sufficient for the purpose, were so great as to render the efforts in this direction but partially success- ful, until the regenerative gas-furnace of Siemens placed in the hands of metallurgists the means of fusing largo bodies of steel on the hearth of a D 294 GEOLOGICAL SURVEY OF CANADA. [50 Martin's steel process, ijts operation, Itci'semer's pro* cess. Hfftton pro- cess. reverberatory. Provided with this, the Messrs. Martin, of Sircuil in France, have succeeded in producing cast steel, in charges of three and four tons at a time, by melting down wrought iron in a bath of cast iron, by what is now known as the Siemens-Martin process. The products thus obtained, attracted much attention at the Paris Exhibition, in 1867, and the process has since been widely adopted in Europe and in the United States; where it was first introduced by Mesprs. Cooper, Hewitt & Co., and is now in successful operation at their woi at Trenton, New Jersey. Beginning with a bath of six hundred weight of pig iron on the hearth, malleable iron, as puddle-bars, for instance, is added, previously heated to whiteness, and rapidly dissolves in the molten cast iron, until, at the end of about four hours, the charge amounts to three tons, and will be found to consist of a soft, nearly decarbonized metal. It is then recarbonized by the addition of from five to eight per cent, of spiegeleiscn (manga- nesian cast iron), as in the Bessemer process, and run in moulds. The bath of molten metal, during the process, is protected by a covering of fused slag or cinder. The furnace-bottom for tliis process is made up of a silicious sand, which must not be quite pure, but contain some alumina or other bases, so that, under the influence of the high temperature, it may harden, without melting, forming an impervious crust, which will resist, for a considerable time, the action of the molten steel. The upper part of the furnace is built of Dinas fire-brick. Attempts have been made to use an admixture of oxyd of iron with the pig metal, in this process ; but it is found that the corro- sive action of the oxyd, at a high temperature, upon the furnace-bed is such as to preclude its employment. The entire cost of a furnace with a capacity of producing three tons of cast steel, with gas-producers, gene- rators, and all the apparatus for moving the ingot-moulds, is, in England, about £500 sterling. This process, it is true, cannot compete with the Bessemer or pneumatic method for the cheap production of cast steel in large quantities ; but while the latter is applicable only to certain fine kinds of cast iron, comparatively free from phosphorus and sulphur, the process in the open hearth permits the employment of other qualities of iron. These, in being reduced, by puddling or otherwise, to the condition of malleable iron, are deprived of the impurities prejudicial to steel, before being added to the iron bath. While, therefore, the Bessemer process will probably remain without a rival for the treatment of the purer cast-irons, the production of steel by the open hearth will perhaps become even more important, because of wider application. The Heaton process, for which so much was claimed as a method for the production of steel from impure cast iron, by the action of nitrate of soda, appears, from the late careful studies of [oa 51] REPORT OF DR. T. STERRY HUNT. 295 The eumatic Lt while I'atively hearth being Ion, are to the ■remain Ition of lortant, bh was Ion, by lies of Gruner, destined to become subsidiary to the production of steel in the open furnace. Gruner concludes that it " can never, from any point of view, become a substitute for the Bessemer and Martin processes. These produce ingots of steel, or homogeneous iron, from pure brands. The Heaton process deals with impure brands, and seeks to convert them into a refined metal, more or less purified, the treatment of which has to be finished in a Siemens furnace." He further declares that the only advantageous way of treating the products of the action of nitrate of soda on cast iron, is to submit them to the Siemens-Martin process. {Annales des Mines for 1869, fifth part.) Mr. Bessemer has very recently made experiments upon the working of his process, under pressure, by which he obtains such an elevation of tem- perature, as, it is expected, will enable him to introduce malleable iron into his converters, and thus effect in them what Martin does upon the open hearth. In the mean time Siemens has, by the aid of his furnace, been able to carry out a part of the original plan of Heath, who, in 1845, ^j,^^^^^f ^ '^''^' proposed to reduce iron ores, by heating them, in small fragments, with charcoal, in a close vessel, as in the methods of Chenot and Clay, and to add the resulting spongy iron to the bath of molten cast iron. The reduc- tion is, by Siemens, effected by a j^lan which combines the indirect and direct methods of Chenot. Above the furnace, and immediately over the bath of molten cast iron, which occupies the hearth, are two large tubes of refractory clay, enclosed in an outer casing, through which the flame from the furnace passes, and allows these tubes, or reduction-chambers, to bo heated, with their contents, to redness. They are charged from the top with finely broken rich ore, through which a current of previously washed and purified carbonic oxyd gas, from the common gas-generator of the furnace, is forced, and reduces the ignited ore to the condition of a metallic sponge of pure iron ; this, descending, is at once dissolved in the molten cast-iron bath, and effects its conversion to steel, precisely as in ^lar tin's plan, where solid malleable iron is made use of. In certain cases, as with very finely divided ores, the reduction is effected by an admixture of about ten per cent, of charcoal, or other carbonaceous matter. Siemens has already manufactured excellent cast steel by this method, and there is no doubt that, in the case where pure oxyds, free from sulphur and phosphorus, can be obtained, the mode of directly producing steel with spongy iron mayj^be advantageously employed . A simple and ingenious process, based, like that of Siemens, on the locWo's patent oirginal suggestion of Heath, has recently been devised and patented by Mr. Robert G. Leckie of ^Montreal. Having found that when finely- divided iron ore, as magnetic iron-sand, was made into lumps with peat, 296 GEOLOGICAL SURVEY OF CANADA. [52 Slomens's rejtonerative furnace. Burning wet fU«l. coal, or other carbonaceous matter, not in excess, and exposed to redness, out of a current of air, there results a nearly pure spongy metaiiic iron, he proposes to obtain iron in this way, and add it to the bath of molten cast iron, in a reverberatory gas-furnace. The ore, agglomerated with the reducing material, is to be placed in one or more large chambers or ovens, in the rear of the hearth, and, when suflBciently heated to eflfect its reduc- tion, is to be added to the bath of molten iron. He expects soon to test, on a working scale, this mode of making cast steel in the open hearth, to which the purified magnetic iron sands of Canada, from their freedom from sulphur and phosphorus, would seem to be peculiarly well adapted. It is one of the great advantages of the Siemens furnace, that by a judicious regulation of the supply of air, and by proportioning it to the gaseous fuel, it is possible to obtain, at will, either an oxydising, a reducing, or a neutral flame ; a point of much importance in the fusion of metals in the open hearth, which was already indicated in Gurlt's specification, as explained on page 46. The employment of gaseous combusibles has been greatly extended since the successful use of the regenerative principle by Siemens. This consists in allowing the heated gases, after combustion in the furnace- chamber, to pass out, downwards, through two chambers packed with fire- bricks, so arranged as to allow a free passage of air between them, to which they impart their heat ; the waste gases passing off into the stack at a temperature seldom above 300'^ Fahrenheit. After an interval of from half-an-hour to an hour, the current is changed, and the gases are led off through another pair of regenerators ; while those which had been heated by the escaping gases are now used to conduct the air and gas for keeping up the combustion ; these passing in through the heated regenerators, have their temperature greatly raised before entering the combustion-chamber. By alternately making each pair of regenerators the channels for the passage of the gases to be burned, and for the waste products of combustion, a very intense temperature is maintained in the chamber, with very little loss of heat. Coal and dry wood have generally been used in the gas-generators, where, by a partial combustion, the solid fuel is converted into com- bustible gases. With wet fuel, a largo amount of steam becomes mingled with the gases, where its presence is very objectionable. This difficulty has, however, been entirely obviated by a system lately devised in Sweden, which may become of great advantage to Canada. I have therefore thought it best to copy from Mr. Abram Hewitt's Report on the Production of Iron and Steel at the Paris Exhibition of 1867, the following account of this valuable invention. This report, published by the United-States Government, contains excellent drawings of the furnace : [5^ 53J REPORT OP DR. T. STERR7 HUNT. "The furnace devised bv T? t j- ' ^^^ designed for the consumption of tu^'l'I ^'t^^'* ^"^ ^^"^^^^^rs, is .«„.„,,, saw-dust or other moist fuel ; an L " . ^' ^'^^' ^'^^ '^ ^^ "" Ssrh;:hre^V-ver;er^^^^^^^^^^^^ -ppiiedfro.he]::trj;;t^^^^^^^^^^^ thus maintained, pass throu.raTnd 7'' ?' P'°^"^'^ ^ '3 condensed. The c^nq th.^ condenser, where all the moisture m .», wi>), «• , ° t^*'^ passes to the hpifmr, r "moisture m the gas With Siemens's regenerators " ^^^t^ng-furnace, which is furnished W, contains one fcrt jt, JT 3 T "' "'"'■S'''' "i"' »"oh wo ■nto a chamber in which, Cm 11^ >''•' ™''°''- '' P'«^<^» »' on'e «.« are fcharged, oroiinglroa t "'^' T" *'^- "' «»« «.« chamber. B3, a;,, ^^ = ^»'' °'h ' m vanous dirccdons, and fillin, " "™ present, with „„ch !f the feaf^ *''"'' f" "'" '«'"' »"" tarry" tricHn ''°»'--««eneralor,, and keDtln\ '"'''"«<''''*' ''><' H-'T:iuot-;t*;eS^^^ » currency, and it ia estimated tl . nt /"r'' '" *™''™' ''^ »""»« »250O 298 GEOLOGICAL SURVEY OF CANADA. [54 Boetius's fur- nace, works, using saw -dust or peat with entire success and great economy. In the lumber regions of Lake Superior it will be found to have a special value, because there is an abundant supply of pig-iron, accessible to the saw-mills on Green Bay and in Michigan, producing enormous quantities of saw-dust, slabs, and waste timber." By the aid of the Lundin furnace, combined with the regenerators of of Siemens, Rinman has succeeded in producing steel by the Martin pro- cess, using only pine saw-dust for fuel- When such results can be obtain- ed with saw-dust, or with ordinary peat, the want of mineral coal need no longer be an obstacle to the development of the metallurgical industry of this country- The gas-furnace of Boetius, which is now used for zinc-smelting, and in many glass-works, in France, is simpler and less expensive than that of Sie- mens- It docs not make use of the regenerative principle, and hence the waste heat can be employed or boilers or for other purposes. In this fur- nace, however, there being no condenser as in the Lundin system, only dry fuel can be made use of. The air which serves to burn the combusti- ble gases in the furnace-chamber, is heated by passing between the walls of the generator and an outer casing, these walls being made very thin, and supported at intervals, by bricks, which arc built both into them and their envelope. This furnace does not enable us to obtain a heat sufficient for the production of cast steel, but is well adapted for puddling and reheating iron, as well as for zinc and glass-works, and is said to economize from 30 to 33 per cent, of the fuel. This description is taken from a paper by Gruncr, professor of metallurgy at the Ecole des Mines of France, which appears, with working-drawings, in the Annates des Mines for 1869, fifth part. The same paper contains, also, descriptions, with drawings, of the Sicmons-iNIartin steel process, besides an account of Pon- sard's experiments, and of the EUershausen process. THE ELLERSHAUSEX PROCESS FOR MALLEABLE IRON. The removal from cast iron of its carbon and silicon, and its conversion MftUoftbie iron, into malleable iron, is chiefly eflfected in two ways : of these the first consists in melting down the pig metal, before the blast, in an open fire known as a hearth-finery or bloomary, somewhat resembling the bloomary hearth used for the {direct process of reduction in the United States. In the second method, the metal is melted and decarbonized in reverberatories, known as puddling-furnaces. In the puddling process the carbon of the iron is removed, partly by the o.xygcn of the air, and partly by that of the oxyd of iron, which, in the form of iron ore, is used for hning, the sides [54 55J "EPOBT OP BB • .. fei»". In one of ^ °''°™'*" '■■"» mall" io lt'''''''''<'''<'"''»™^atio, ""•- « four d^; ZZ1' *''•' "°«""os, a n; ,t "' '" P"'''^" fflalloable ion. i„ .u ° '" <l«««rl»n«d, and eha„4d « ,'"«' '^ °<" '» «on 'lootrt'i °'-*»'« « i«n trr'r""""' f^"«-^ ™ found ,hat iritrf *'' '""dheat A fj""7 "P»«d for ireldcd to^etliBr , ® °' "■"" «re decarhnn- ""'' °f «'« time it -j^« Of C;!;:^' ^-"' '" f-. oor:t: • -tx^'' r^''"^ % another process tl ""^^^'^^Wo iron by the racked hi boxes of 'r? '''''^^^^ to three-fl! i" ^"""-^^'^ "method . ;:"^e the nietal is found ^e', ^^ ^^^ «^^«ral weefs . H. ' ^'''''^' '^ ^^'^^Uie endeavour !'^^' "^^^<" experir^entf ^ ^'''''' bear- „, "^«ture, and thu !b -^^^ "^^^«''on of subseo I?, '"'^^ ^^ "•<>" with ^'ith thatofwrl ^'^'ainujgcast steel tf '^"'^"''j meltin- do,vn I , Ho fonnd\!rrzfr"' *»2 '^'i" '""" ™'™'»"^ '"»:: '.oaled on iho heard, 'p ° """PO"'" m4ts of „ , n.elal ,™ rapid !l;'""<=*"alorj, fc;'?';? ''"'' P'^aotal, wl,e„ P^'l-od b^ thJiY^r 't'' '»'» "-"^ of a?'?"'' »"'•' to taken 300 GEOLOGICAL SURVEY OF CANADA. [56 Theory of the proceEe. effecting the conversion of the metal, but subsequent experiments have shown that by reducing the proportion of ore much below that required, by theory, to effect the change ; and even by replacing a portion of the ore by powdered charcoal, whose effect would seem to be the reverse of oxy- dizing and decarbonizing, as good results were obtained as before. In the pig-bloom, as the aggregate of pig metal and ore is termed, the iron is much subdivided, being partly in graios, and partly enveloping the granules of iron ore ; the whole forming a somewhat porous aggregate, which is pervi- ous to air, and thus offers a great extent of surface to its oxydizing action, as well as to the action of the intermingled oxyd of iron. Where an admix- ture of charcoal is used, it would soon be destroyed by combustion, and by the action of the accompanying iron-oxyd, and the mass rendered still more permeable to the air ; so that the finely-divided white cast iron of the pig- bloom becomes rapidly decarbonized under the joint influences of the oxygen of the air and that of the ore. The ore, being hi part reduced to the metallic state by the carbon and silicon of the cast iron, tends to make the loss of iron less than in the puddling process. In this view, the Ellershausen method unites the reactions of the process for malleable castings, and the Welsh process above described, where oxyds of iron are the decarbonizing agent, with that of Tunner, in which the de carbonization is effected by the oxygen of the air. If we suppose the oxygen of the mingled iron ore to be the sole decarbon- izing and purifying agent, the reaction would be as follows : the carbon of the pig iron, with the oxyd of iron, would give rise to metallic iron and car- bonic-oxyd gas ; while the silicon, which the crude metal always contains, in variable quantities, would reduce another portion of the oxyd, liberating metallic iron, and forming silicic acid. This, in its turn, would unite with a portion of unreduced oxyd of iron, to form a fusible silicate or slag, of the composition already referred to on page 282. If we take the magnetic oxyd of iron, the reaction with carbon would be represented by FcjO, +4C = 3Fe + 4CO, while with silicon we should have FC3O, +Si = Fe + SiO,,2FeO. The above equations lead to the following results for each unit of car- bon and silicon in the pig iron : 1 carbon requires 4'83 magnetic oxyd, and gives iron 36, carbonic oxyd 2-33 1 silicon " 8-28 " " " " " 2'0, silicate of iron'_7-28 Thus a pig iron holding, for example, 95*00 per cent. of|iron, 4-00 per cent, of carbon, and l-OO of silicon would require, 4 X 4-83= 19'32 of magnetic oxyd. 1 X 8-28= 8-28 " " " or. 27-60 " 67:i REPORT OF DR. T. STERRT HUNT. 301 and should yield 16 parts of reduced iron, and 7 "28 of silicate of iron. In the case of some pig irons, which, in addition to 4-0 or 4-5 per cent, of carbon, contain 2*0, or even 2-5 per cent of silicon, the quantity of mag- netic oxyd required, according to tae above formulas, would be greatly increased. In the trials on a large scale, for the production of malleable iron by the Ellershausen method, at Pittsburg, Dr. Otto Wuthmade care- wuth-aanai. ful analyses of the pig metal, and the resulting products, both iron and slag. From these analyses it appears that when 100 parts of a metal, holding over 1-0 per cent, of silicon and 4-2 per cent, of carbon, were mi.xed with from 28 to 30 parts of magnetic or hematitic iron ore, and treated as above described, the silicon, and nine-tenths of the carbon were removed, together with most of the sulphur and phosphorus. At the same time the resulting slag was much richer in iron than that obtained in puddling the same iron, or, indeed, than most slags from the puddling-fumacc. It contained an amount of iron equal to not less than 64-7 per cent, of metal, and but 8'95 per cent, of silica, while the saturated silicate of iron, whose for- mula is given above, contains but 54-9 per cent, of iron, and 29-4 per cent of silicon. The highly basic slag from the Ellershausen process, as anal- yzed by Dr. Wuth, has thus a composition corresponding to a mixture of about 30 per cent, of a saturated silicate of protoxyd of iron, (with small portions of lime, magnesia, and alumina,) and 70 per cent, of magnetic oxyd of iron. From this it appears that a large part of the ore added to the pig metal is not consumed, but passes off in the slag ; and it would seem that, in thi.s case, the principal action of the oxyd of iron had been the removal of the oxydized silicon. Each unit of silicon furnishes by its oxydation an amount suicious iron, of silica which requires at least four units of iron, in the state of protoxyd, for its conversion into the ordinary fusible silicate of iron. All of this oxyd of iron, in the ordinary puddling-process, except so far as furnished by the fettling, must be derived from the oxydation of the metal, and hence the great waste with highly siliciferous cast iron in the puddling-furnace. For such irons, therefore, the Ellershausen process would seem to be especially adapted. Were the conversion of the iron to take place according to the formulas already given, solely by the action of the oxyd of iron on the carbon ai: 1 silicon of the pig metal, 100 parts of this, having the composition above assigned, should yield theoretically, supposing no subsequent loss of iron xiicoryoftuo by oxydation, or otherwise. 111 parts of pure iron ; since to the 95 parts ^''^^^^^' present in the pig metal, would be added 16 parts reduced from the oxyd, hy the carbon and silicon. In practice, however, the gain is much less than this, leading to the conclusion that a part of the carbon is oxydized by atmospheric oxygen, while much of the added iron-oxyd must escape 302 GEOLOGICAL SURVEY OF CANADA. [58 unreduced, in the slags, as we have seen is really the case.- According to Dr. Wuth, the result of the treatment of nearly 4000 tons of iron by the Ellershauscn method, as above described, with about 28 per cent, of o.Kyd of iron, showed a gain of not quite 5 per cent, on the weight of the pi;; iron employed. These conclucions are confirmed by recent results of the iron-works of Messrs. Burden, at Troy, New York, where the Ellershauscn process has been found to give satisfactory results, with 15 per cent of magnetic iron ore, although the quality of the product was improved when 20 per cent, of ore was used. Analyses of the pig metal, the ore, and the products, in such trials will be most important as serving to shed farther light on this new process. Mean- Suptirostions for while the following suggestions with regard to it seem warranted bv the practice. ; ° ~° ° ^ facts Ijefore us. 1st. The ore used should be as free as possible from im- purities. Silicious matters, by uniting directly with the oxyd of iron, occa- sion a large loss of ore ; while lime, magnesia and alumina-compounds, not only increase of the bulk of slag, but render it pasty and difficult to be removed from the iron. 2nd. The ore should be finely divided, inasmuch as more surface will thus be presented to the iron. In the working of the process at Pittsburg, much of the ore added was in coarse grains, which escaping, dissolved in the slag, but otherwise unchanged, caused this to be, as we have seen, extremely rich in oxyd of iron. The coarse grains, it may be supposed, serve however to give to the aggregate that mechani- cal condition which is favorable to the proper working of the process, a ri>sult which would probably be equally well secured by the admixture of a portion of charcoal ; an experiment, which I am informed, has already been successfully tried at Pittsburg. The use of a grc:itly reduced proportion of finely divided and very pure ore, together with a portion of coarsely ground charcoal, would therefore seem to promise the best and most economical results with the Ellershauscn process. Rich hematite, free from Choice of oriv. siUca, or magnetite, previously calcined, and if necessary, purified, after crushing, by theaid of a magnetic machine, should be tried. The magnetic portion of the fine iron sands from the lower St. Lawrence would proba- bly yield excellent results in this process. Sonv experiments made at Pittsburg, in which the purified iron-sn . place of the ordinary ores, are said to have given r\ sa[it>i'. n. The ores used in the trials which gave the \ by J Wuth, were, however, the magnetite of L^ko Chamj .i, witii ae hematite from Missouri. From what has been said, it will be < ;ident that the supply of air in the furnace should be as abundant as in >' process of puddling, and that a reducing or feebly oxydizing atmosphere therein, would e or greatly modify the conditions of the Ellershausen process, or lead to lure. t r 01 ur COi anc fun ed, and It fequi treati of ih t»einjj about masses of the ; also o\ ^'t is cj£ that the from ihQ apparent the sulpi ^llershau of the rep eupplemen Osborn'g j) The Eji, Messrs. Shi States ; and fined to gen causes, amoB I PartiaJ foiiuj preceding pa [58 [cording to I'O" hy the Jt. of o.vyd |of the pig Ji- works of ■'•ocess lias I'letic iron 1 per cent. als will he K Alean- M by the ft'om im- P", occa- "icls, not lit to he iiasnmcJi ^^' of the '3, which this to grains, aechani- J'ocess, a ;ure of a ■dy been >portion oarseJy i most Je from , after gnetic )roha- iJe at iiiary id in Jver, 59] ce,avf '"''"'* arcs. '7' '"""a.c I" "'"'^*'o„ f ° '^'''o 60;^; *° ""'I fWter '"'■'' 'o ko „? '^* 'eo(i, J "*«*W *° '"■"'OSS is ' °':"«'«l iaV'""^^. efe«" ^''*'o„ '„ ''"■''"««. *, f «»»• ^V ^ "'«, «,„' y,'"' Ellorsi^yj/ oj- J-^n ' "'' "ith Z>r t '"'"<^ar, 1 ""°»arv ^cedin^ J''' the prohr,i.7y''^^^sies Jj" A«« ^een .„../? J"^5-e, i., ., "'^ ^^^^"^ 304 GEOLOGICAL SURVEY OF CANADA. [60 Granulated iron. Numerous patent-claims, from that of John Wood, in 1761, down to the present time, have been based upon the use of granulated or pulverized cast iron for the production of steel or malleable iron. The iron is granu- lated by beating in large mortars, when heated nearly to its melting point, or by causing it to fall into water, through the air, or upon a rapidly- revolving disk, from which it is thrown off by centrifugal force. The grains of iron, more or less oxydized at the surface, are directed to be con- veyed to a furnace, and there formed into lumps for the rolls or hammer ; or else mixed with oxyd of iron, and exposed to heat in a furnace, (or in close vessels) whereby a malleable iron, fit for the manufacture of steel, is obtained. See, among others the specification of Bousfield, in 1857, No. 3082, and that of Morgans, in 1865, No. 806, of the British Patent Ofiice. In so far as these propose to work in the open furnace, they differ from the old method of Wood, and the Welsh process, already described, page 299, and approach to the conditions attained in the EUershausen process. Excellent results have recently been obtained by Mr. Hewitt at Ringwood, New Jersey, by mixing the granulated cast iron, with iron ore in grains, and exposing the mixture to heat on the hearth of a reveberatory ; when decarbonization, and conversion to malleable iron takes place, as in EUershausen's method, without fusion. ■P"*"P? [60 m to the ulverized is granu- ag point, rapidlj- se. The ) be con- laminer ; 36, (or in ' steel, is 557, No. I Patent ley differ jscribed, I'shausen [ewitt at iron ore eratory ; e, as in 'm