Search Documents

Executive A R LEDUC, Chairman ARTHUR S DWIGHT R V NORRIS EDWIN LUDLOW E L DEGOLYER Membership J V W REYNDERS, Chairman E L DEGOLYER R M RAYMOND EDGAR RICKARD W Y WESTERVELT Finance WALTER H ALDRIDGE, Chairman GALEN H CLEVENGER J V W REYNDERS Library Chairman JOHN H JANEWAY GEORGE C STONES ROBERT LINTON ALEXANDER C HUMPHRLYS, - BRADLEY STOUGHTON

Jan 1, 1923

By Aloke Verma, R. Singh, S. Jana, C. Sawmliana, P. Pal Roy

On July 21, 2004, a trial blast using electronic detonators was conducted in India for the first time in an opencast mine at Pundi-E Quarry of TISCO, West Bokaro. Orica-IEL conducted the test using electronic detonators. To provide necessary technical feedback on wider applicability, CMRI scientists also participated in the trial blast.

Jan 1, 2006

By Thomas L. Richardson, David R. Chapman, Gary W. Gilbert

The two contracts of the Big Walnut Augmentation/Rickenbacker Interceptor(BWARI) sewer project for the City of Columbus, Ohio total nearly 35,000 feet of 14and 12-foot diameter tunnel. EPB TBMs will install precast concrete segmental lining through difficult glacial soils ranging from clean granular to hard till with high groundwater and boulders, crossing roads, railroads, and a large stream. Construction of 12 shafts and a large Interconnect Structure involving major ground stabilization work is required. The paper will review design of major project elements and present an update on construction progress.

Jan 1, 2005

Neutralization with lime is a simple, cheap, and effective means of treating acid raffinates from uranium processing to' raise the pH, reduce the concentrations of dissolved salts (including heavy metal impurities), and remove amine by adsorption. Two-stage neutralization, using limestone in the first stage at approximately pH 4 and lime in the second, lIlay be advantageous.Lime neutralization of mffinates in the pH range 7 to 8· 5 generally reduced the radium ·concentration to 10 to 200 pCi 1-1, but ·there was considerable scatter in the experimental data. The radium concentration was further reduced to less than 3 pCi 1-1 by co-precipitation of barium (radium) sulphate with 10 mg Ba (as barium chloride) per litre of raffinate. Adsorption of radium on barytes was -also effective in reducing the radium concentration in neutralized raffinate to less than 3 pCi 1-1.The radium concentration of pit water was readily reduced to less than 3 pCi 1-1 when pit water containing 0·01 M sulphate as ammonium, calcium, or sodium sulphate was treated with 20 mg Ba per litre of pit water.Adsorption on barytes was also effective in removing radium from pit water containing 0·03 M sulphate.

Jan 1, 1975

Sluiceboxes can provide a much higher concentration ratio than most other gravity concentrators (up to 50 000:1) at very high overall placer gold recoveries (greater than 99 per cent). They are also very reliable, inexpensive and simple to operate. This combination is very difficult to beat and explains why the sluicebox is still the most important placer gold concentrator in northwestern Canada, Alaska and the rest of the world. A sluicebox is a rectangular fl ume containing riffles on matting, through which a dilute (<12 per cent) slurry of water and alluvial gravel flows. Sluiceboxes are actually centrifugal concentrators with settling velocity playing a minor role in gold recovery. Due to their higher density, gold particles tend to segregate to the bottom of the slurry flow where they form a streamline that is diverted by a low pressure zone into a riffle. Under ideal conditions, this ribbon of slurry will be overturned as it flows down the rear of the following riffle and will continue fl owing in a circular path to form a vortex. At the bottom of this vortex, centrifugal and gravitational forces combine to drive gold particles into or beside the matting. To remove the gold concentrates, sluiceboxes are shut down and the riffles and matting are taken apart and cleaned. The slurry velocity provides the energy that powers the vortex. If the velocity of the slurry is reduced through overloading with solids, insufficient water flows or shallow gradients it may not sustain a full size vortex. If the riffles are too close, too far apart, too tall, or if there is not enough energy available to the vortex, the vortex will not be formed properly and gold recovery will be reduced. If the slurry velocity is too high, extreme turbulence and the resulting scouring will also cause gold losses. Testing sluiceboxes with conventional sampling and evaluation techniques is very costly, time consuming and problematic. Most placer gold ores contain a very small number of gold particles in a large volume of pay gravels. Sluiceboxes can lose coarse gold particles. The presence or absence of one of these particles in a tailings sample can lead to high unpredictable errors (nugget effect) even when large sample volumes are processed with care. Every time conventional samples are upgraded, additional errors are introduced due to the inefficiency of recovery equipment. Nuclear tracer tests are more accurate, faster, cheaper and safer than conventional sampling. In 1989 through 2002, the recovery efficiency of dozens of sluicing systems was determined by mixing radioactive gold particles (tracers) into the feed streams of dozens of placer mines in the Yukon Territory, British Columbia, Alaska and Guyana. Four distinct sizes of nuclear tracers were used and their recovery was related to the design and operational characteristics of the individual sluiceboxes and their pay gravels. At clean up, scintillometers were used to detect the very low level X-ray and gamma ray radiation emitted by these tracers and locate them in the sluice runs. The effectiveness of any given section of riffles was easily diagnosed by the presence or absence of tracers. At every mine the gravel feed rates, water flows, equipment dimensions and riffle performance were measured. After the gold tracers were removed from the fi nal concentrate and counted, the concentrate was sieved and weighed. These gold recovery, weight and sieve data were used to calculate the quantities and size distributions of the gold particles in the original pay gravels and those lost in the tailings. The expired tracers were stored in a lead lined container until their radioactivity was near background levels (about two months). The standard errors from these radiotracer tests were estimated from binomial probability theory with the equation: SE = {(n + p + q) ^ 0.5}/n. Where n was the total number of radiotracers added, p is the proportion recovered and q was the proportion lost. Each overall recovery estimate would be within one standard error of the true recovery value (14 times out of 20) and almost always within two standard errors of the true recovery (19 times out of 20). The maximum standard error with 100 tracers was five per cent and occurred when the recovery approached 50 per cent. With higher recoveries, standard errors usually ranged from one to two per cent.

Aug 1, 2010

By S Henderson, J V. Lawless

The worldwide association of low-sulfidation epithermal gold mineralisation, bimodal rhyolitic-basaltic volcanism and rifting is confirmed in New Zealand. In the currently productive Coromandel epithermal-gold province reported Ar-Ar dating of adularia from low sulfidation deposits falls into two groups at 6 and 7 Ma. Bimodal rhyolitic-basaltic volcanism has been reported to be K-Ar dated at 9 to 7 Ma and 6 to 5.5 Ma. Epithermal gold mineralisation in Northland is spatially associated with bimodal basaltic-rhyolitic volcanism, with rifting cogenetic with that in the CVZ. The Taupo Volcanic zone (TVZ) is sufficiently young that the presence of a segmented rift with recent bimodal rhyolitic-basaltic volcanism can be genetically associated with epithermal gold mineralisation, which occurs as both fossil occurrences and in active geothermal fields. This reinforces the known potential prospectiveness of the Coromandel area and increases that of Northland and the Taupo Volcanic Zone.

Jan 1, 2005

By F. Hassani, K. Moozar, J. Ouellet, P. Gauthier

"This paper describes a sudden spill of waste rock used for backfilling from a waste pass at an underground mine in Canada, thus causing a major failure/accident. The mine operator immediately launched a detailed investigation to understand the causes of this accident and develop procedures to prevent any recurrence of such events in the future. The authors from McGill University were invited to collaborate with the engineers at the mine to examine this accident. The events leading to the accident were studied, appropriate measurements were taken and samples of the waste material were collected for further testing and analysis.The field visit and the interview of the witnesses allowed the detailed documentation of the events. The sequence of events related to the incident, along with the experimental results obtained in the course of the investigation, are presented. The waste material was characterized in the laboratory with regard to its behaviour in the presence of water. This information is discussed and the failure mechanism involved in the accident is identified. Finally, the full recommendations as well as the immediate operational measures implemented by the mine to prevent the future occurrence of such events are presented and discussed."

Jan 1, 2004

IN the recent monograph on Broken Hill, N.S.W., by E. C. Andrews,* two independent petrological reports appear as Appendices 1 and n, by W. R. Browne and F. L. Stillwell respectively. In these considerable differences of opinion occur. Andrews contributes to some of the questions at issue, but no analytical review of the differences appears in the body of the report. Their development may invite question, and the following notes have been prepared partly in explanation and partly as a summary of some of the divergences. The soope of Browne's report extends over the whole district of Broken Hill, while Stillwell's deals with the rocks within 25 chains of the main lode. The more difficult and important rock-types occur in both areas, raising similar questions. In both cases observations extended beyond the limits of the prescribed areas. Broadly, Browne faces his problem as a study of a series of primary igneous rooks, acid and basio, whioh are intrusive into a series of sillimanite gneisses. The development of the sillimanite gneisses is regarded as belonging to an earlier period than the intrusion of the igneous rooks, while the gneissio structures of the latter are features which are incidental to their orystallization, and are not acquired subsequent to their consolidation. He finally produoes a picture in which his views of igneous injection and differentiation takes a prominent place. One direct result is the claim that three separate periods of metamorphism have oocurred in the Broken Hill area. Stillwell primarily examines the rooks as schists and gneisses since their present dominating features have been acquired during metamorphism. The effort is made to understand suoh differences as might arise from differing conditions of metamorphism upon rocks of similar and different oonstitution. The aim is to divest the rocks of those oharaoters which are due to metamorphism before approaching the questions of intrusive relatio'nships between igneous rocks andsediments.

Jan 1, 1924

This paper describes the methods of constructing timber barricades against pillars of ore in the South Mine, property of Broken Hill South Limited, Broken Hill, N.S.W., and discusses the behaviour of the various types of barricade alongside trucked sand filling and hydraulic stowed sand filling.Since the introduction of hydraulic stowing, cut and fill and overhand square letting methods have been used in mining the pillars. Such methods could not be used where the pillars had failed badly. The filling used is partially deslimed concentrating millresidue containing among other gangue minerals, calcite and a small amount of pyrrhotite and pyrite.The author is indebted to the mine superintendent of Broken Hill South Limited for permission to publish this paper.OLD METHODS OF CONSTRUCTING BARRICADES.Square Set Stopes.Before filling each stope the outer row of legs was lined inside with 10 in. x 2 in. boards spaced 2 in. apart. It was usual to tie the row of legs and caps together with timber dogs. As the space between the barricade and ground varied from nil to several feet, it was general practice to fill behind the lining boards. No provision was made to retain this sand during subsequent pillar stoping, consequently a run of sand right up the barricade could cause adilution of ore and serIOUS dIstortion of the barricade.

Jan 1, 1946

By R Handfield, G Kary, Frieda Project

Recent exploration by the Cyprus Amax-Highlands Pacific-OMRD Joint Venture at the Frieda River project, PNG has demonstrated the extent of igneous phases, alteration and mineralisation, and the relationships between the various mineralisation styles in this major Cu-Au porphyry-epithermal system. The project area is in the obducted volcano-sedimentary section of the New Guinea mobile belt 78 km NE of Ok Tedi. It is between two major NNE dipping faults interpreted to have undergone thrusting then sinistral shear in the Late Miocene to Recent. The Frieda River Igneous Complex is a 15 km +7 km ovoid area with a central pod of basaltic andesite volcanics with cogenetic diorite to quartz diorite subvolcanic intrusions, and flanking clastic sediments. A series of late-stage, hornblende quartz diorite plugs ring the main igneous mass. The igneous rocks have a normal K calc-alkaline character with a very limited compositional variation. The intrusives are exposed on the eastern flank of the complex and have five mappable units and numerous defined phases. The southeastern quarter of the complex is host to a cluster of at least seven stockwork porphyry copper and skarn-hosted deposits within an area of 4.5 +3.5 km. The Horse-Ivaal and Koki deposits have identified resources totalling 804 mt @ 0.46 per cent Cu and 0.3 ppm Au, and the other deposits are in various stages of exploration. All these deposits are related to the Horse Microdiorite phase of the Complex and are outlined by a zone of potassic-propylitic-SCC alteration. Local overprinting argillic and advanced argillic alteration is related to a separate, post-porphyry mineralisation intrusive phase. The recent emphasis has been on the identification of higher-grade zones using ideas of structural control and specific intrusive-alteration-mineralisation relationships within the known resource areas. An elongate area 13 km x 4 km in the centre of the Complex has advanced argillic alteration in the volcanic rocks. This alteration zone hosts the Nena deposit (52.8 mt @ two per cent Cu and 0.7 ppm Au) and approximately 20 prospects of high sulphidation type. These deposits are distributed along two prominent NW-trending siliceous ridges interpreted as regional scale structures. Nena is a pod of vein and replacement sulphides hosted in a tectonically controlled multiphase breccia that overprints the advanced argillic alteration. The Nena deposit is 4 km from the nearest of the major porphyry occurrences. It is underlain by basemetal veins and minor porphyry mineralisation that are related to the Flimtem Trachyandesite intrusive phase. The recent search for additional Nena-type ore has focussed on the siliceous ridges using detailed structural and geochemical ore control criteria. Widespread, but poorly mineralised, basemetal-bearing carbonate and sulphate veins form a ring around the mineralised area. These veins have sericitic selvages, cut the porphyry mineralisation, but are cut by and locally underlie the silica-alunite phase of the advanced argillic alteration. The current model is that there are discrete intrusive phases responsible for each of major hydrothermal events at Frieda. These include from oldest to youngest:the Koki phase has no related alteration or mineralisation; the dominant Frieda phase has the hornfels, early skarn and minor magnetite-chalcopyrite mineralisation; the Horse phase has the bulk of porphyry copper and skarn-hosted mineralisation and minor basemetal veins; the peripheral Knob phase is barren; and the Flimtem phase has the high sulphidation epithermal deposits and minor porphyry and vein mineralisation. The intrusive phases span the interval 17 to 11.5 Ma and are co-magmatic. They represent discrete batches off the parent magma, rather than simple fractionation products.

Jan 1, 1999

By A. T. Haile

Detailed monitoring of the response of the rock mass between rockbolt reinforcement units subjected to a simulated seismic source has been successfully conducted. This work has shown the increase in amplification of the Peak Particle Velocity (PPV) with distance from a rock bolt unit, and, a minimum PPV for rockburst damage for the given site characteristics. This understanding, at this site, will allow the design of a suitable rock bolt spacing to prevent unravelling of the rock mass between the rock bolt units for a given level of seismicity. Or, alternatively, an estimation of the requirement for a suitable fabric support system, for the anticipated level of seismicity, can be made.

Jan 1, 2001

By M. Baumann

Introduction Gold Quarry is one of the largest gold deposits in Nevada, located along the Carlin Trend, approximately seven miles north of Carlin, NV. Owned and operated by Newmont Mining Corporation, as of December 31, 2005, Gold Quarry had produced over 13 million ounces of gold. Multiple simulations were constructed for gold and metallurgical variables using the Stepwise Conditional Transformation method. Analyses of these simulations have been used for a variety of purposes, including an assessment of uncertainty with various ore types. This assessment provides insight for improved modeling and reconciliation. Geological Framework Gold Quarry is a large, low-grade disseminated gold deposit hosted in Silurian-Devonian aged sediments at the Chukar/Alunite-Good Hope fault intersection. Gold bearing hydrothermal fluids appear to have migrated upwards along the Chukar/Alunite and Deep Sulfide Feeder faults and then moved outward along favorable stratigraphic horizons and structural intersections.

Jan 1, 2007

By Richard Peaker, Jens Lichter, Alexander V. Potapov

"The recent combination of Computational Fluid Dynamics (CFD) and Discrete Element Modeling (DEM) can provide considerable insight into the performance of flotation cells. One area of interest from a fundamental equipment design perspective has been the influence of flotation cell size and the inflow rate on the metallurgical performance of the cell. One of the specific design objectives when increasing the size of flotation cells, or increasing the feed flow to the flotation cell, is to ensure that the feed stream flowing into the cell does not have an opportunity to bypass the mechanism.This paper describes the initial modeling efforts to quantify the inter-relationship of cell size, inflow rates, and feed and discharge box placement, on the retention time distribution of solids in the flotation cell, and the fraction of particles that bypass the mechanism on a size by size basis.INTRODUCTIONThe great majority of newly installed cells in mineral processing plants world-wide are of the cylindrical tank design and this tank shape offers a number of hydraulic advantages over the older style of square cells. Cylindrical cells give symmetrical flow patterns and by careful design of the mechanism and tank internals, the flow patterns generated can concentrate the more intense mixing in the lower portion of the cell to promote good solids suspension and particlebubble contact. Simultaneously, cylindrical tank design gives a more quiescent upper zone for particle-bubble transport into the froth zone. These symmetrical flow patterns are naturally disturbed in normal operation by the slurry inflow and outflow from the individual cells. The challenge for flotation cell designers is to determine the magnitude of these flow disturbances, and to assess whether or when the inflow and outflow rates have a sufficient impact on the hydraulic flow patterns so as to potentially affect the metallurgical performance of the flotation cell."

Jan 1, 2007

By Stephen C. Tadolini, Raymond A. Mazzoni

The International Ground Control Conference in Mining has always provided an open forum for the publication, presentation, discussion, and often heated debate on roof bolting systems mostly with attention to how, when, and why they work. During the last 24 Conferences at least 170 papers have been presented by International Experts on these “simple” devices that range in length from 30-inches to 20 ft and from 5/8 to 2 inches in diameter. Roof bolts are primarily mechanically anchored, glued, cemented, or driven. Bolts are placed vertically, angled, or tied together with special fixtures (trusses and slings). They have been termed active, passive, stick-slip, and several other descriptive acronyms or mining slang expressions. This retrospective paper will present the changes in bolt types and usage, point out some of the biggest myths and hail the most significant advances (according to the authors’ opinions of course).

By W. Biermann

All industrial processes are economically driven, resulting in the ever increasing need for process optimization in order to stay competitive and profitable. Numerous models has been published depicting the optimization of the raw material usage in an arc furnace, as well as the thermal and electrical efficiency; however, to obtain a holistic view of the cost-effectiveness of the process one needs to expand the model to include the cost of all the operational variables. This project therefore aimed to not only include the raw material feeds and power consumption as input variables, but also the labour, maintenance, and environmental costs to generate an all-inclusive financial balance. In the end, the model was able to mathematically optimize the ferrochrome (FeCr) production process through evaluating the effect of a specified combination of input parameters. It could therefore be used as a tool to aid in managerial decision-making; to appraise and correspondingly adjust certain operational variables in order to achieve a maximum return on investment. This paper fully describes the reasoning behind the choice of the essential model components, the assumptions made during the models development, the calculation approach, as well as the advantages, limitations, possible applications, and areas for expansion of the final model. Through running simulations with plant data, it was found that the accuracy of final model remained fairly constant when applied to stable furnace conditions, but was met with a number of limitations when applied to more complex situations. These limitations could, however, be overcome through further development and expansion of the existing model to include a larger number of process variables.

Jan 1, 2012

The mineralogy and microstructure of mattes from the nickel flash smelter at Kalgoorlie, Western Australia, have been studied. The flash furnace matte consisted mainly of an intimate mixture of pentlandite and heazlewoodite with some nickel metal. The electric furnace matte was mainly an iron-nickel alloy with subsidiary pentlandite while the converter matte consisted mainly of heazlewoodite with crystals of nickel metal.

Jan 1, 1977

By H. J. McQueen

ORDER OF PRESENTATION ?INTRODUCTION TO THE VICTORIA BRIDGE ?REASONS FOR A BRIDGE AT MONTREAL ?RAILROAD BRIDGE REQUIREMENTS ?IRON PRODUCTION CANADA / WORLD AL ?METALLURGICAL INDUSTRY CANADA ?FAIRBAIRN, STEPHENSON SHIP PLATE-RIB BOX GIRDER ?VICTORIA BRIDGE (PREFABRICATED BRITAIN) ?IRON ORE SITUATION THE VICTORIA BRIDGE 1859, TUBULAR -WROUGHT IRON* 1898, TRUSS ?STEEL* (*HOT RIVETED) 1957, SPUR, LIFT SPANS ??DUCTILE WROUGHT IRON WITHSTANDS FATIGUE ?TO KEEP THE TRAINS ON TIME & ON TRACK? ??IMMOVABLE SHARP PIERS SPLIT ?THE IRRESISTIBLE ICE, STOPPED CITY FLOODS? ? ?********* ??111 YEARS-YOUNG STEEL TRUSS, (1898): ?STILL VITAL SEA-T0-SEA RAIL LINK? ? ?********* ??51 YEARS HAVE PASSED SINCE SEAWAY (1958) ?SWITCHED BRIDGE INTO DOUBLE LIFT SPANS?

May 1, 2011

By J. Wagner, P. Eger, G. Melchert

In 1991, LTV Steel Mining Company decided to close an open pit taconite mine in northeastern Minnesota, using a passive treatment approach consisting of limiting infiltration into the stockpiles and wetland treatment to remove metals. Over 50 million metric tons of sulfide-containing waste had been stockpiled adjacent to the mine during its 30 years of operation. Drainage from the stockpiles contained elevated levels of copper, nickel, cobalt and zinc. Nickel is the major trace metal in the drainages, and prior to closure annual median concentrations ranged from 1.5-50 mg/L. Copper, cobalt and zinc are also present but are generally less than 5% of the nickel values. Median pH ranges from 5.0 to 7.5, but most of the stockpile drainages have pH greater than 6.5. Based on the chemical composition of each stockpile, a cover material was selected. The higher the potential a stockpile had to produce acid drainage, the lower the permeability of the capping material required. Covers ranged from overburden soil removed at the mine to a flexible plastic liner. Predictions of the reduction in infiltration ranged from 40% for the native soil to over 90% for the plastic liner. Five constructed wetlands have been installed since 1992, and have removed 60-90% of the nickel in the drainages. Total capital costs for all the infiltration reduction and wetlands exceeded $6.5 million, but maintenance costs are less than 1% of those for an active treatment plant. Since mine drainage problems can continue for over a hundred years, the lower annual operating costs should pay for the construction of the wetland treatment systems within seven years.

Jan 1, 1999

By R Castedo, L. M. Lopez, J. A. Sanchidrian

In 2014 the authors presented results from 13 tests made with two blasting seismographs on a vibration exciter. They measured coupling transmissibility as function of frequency (i.e. ratio of the velocity of the sensor mount to the velocity of the ground motion) from 16 to 200 Hz at one or two vibrations levels (5 mm/s or 0.2 in/s, and 20 mm/s or 0.79 in/s) with mounts freely placed, sandbagged, and anchored to a granite slab firmly attached to the plate of a vibration exciter. As the memory capacity of one of the seismographs prevented to cover low frequencies, e.g. 4-16 Hz, a new measuring campaign has been recently made to extend the study down to 2 Hz. In this campaign, two more techniques, in which the sensor mounts were fixed with thermal adhesive or a combination of duct tape and gypsum plaster, have been also investigated from 2 to 200 Hz at 5 and 20 mm/s. Free and sandbagged mounts lead to a weak coupling of the mounts to rock, so that transmissibility depends on the characteristics of the imposed rock motion (peak velocity and frequency). Measurements are accurate only at low frequencies (below 17 Hz for free mounts and 10 Hz for sandbagged) in the tested conditions (velocities up to 20 mm/s). Above these frequencies, ground motion in the measurement can be modified by a factor of 0.16 to 1.22. Anchored, glued and plastered mounts achieve a stiff rock-to-ground coupling, ensuring consistent measurements for the frequencies commonly found in blasting independently of the vibration level and the mount characteristics

Jan 1, 2015

By Raymond M. Howe

Additional data have been secured on the disintegration of furnace linings and the spalling of stove brick. The first article on the disintegration of furnace linings appears to have been written by John Pattinson in 1876.l The furnace involved was 85 ft. (25.0 m.) higher and 27 ft. (8.2 m.) in diameter at the boshes; after being in blast five years, the top brick became contaminated with carbon and disintegrated. An examination of the lining showed that the discolored, disintegrated brick were in a zone found from 35 to 67 ft. (10.6 to 20.4 m.) above the hearth; immediately below, in a zone extending from 22 to 40 ft. (6.7 to 12.1 m.) above the hearth, the brick were hard and porcelain-like in appearance. Analyses of samples taken from the different sections are given in Table I. The water-soluble portion found in the sample taken 39 ft. (11.8 m.) above the hearth had a composition approximately as follows: silica, 0.50 per cent.; lime, 0.04 per cent.; potash, 0.78 per cent.; soda, 0.20 per cent.; potassium chloride, 6.19 per cent.; moisture, 1.03 per cent.; total, 8.74 per cent. The carbon Table 1.—-Analyses of Samples Taken from Diferent Sections of Furnace . I ii In Iv v Section 67 Section 45 Seotion 39 Section of Original ft Above ft above ft above vitrified Lining Hearth Hearth Hearth Portion Per Cent. per Cent per Cent per cent. per Cent Silica........................... 55.70 55.60 55.67 47.60 37.07, Alumina........................ 35.50 35.40 30.00 26.47 23.18 Ferric oxide..................... 3.00 1.32 0.96 Lime........................... 0.23 0.06 0.17 Magnesia....................... 1.22 0.95 0.86 0.00 Potash......................... 2.60 2.10 5.30 7.54 21.59 Soda.........,.................. 0.45 0.42 1.67 0-76 6.28 Ca1bon ......................... 2.05 3.24 5.97 Carbon dioxide.................................. 2.. 17 Water...................................... 7.70 Water-soluble........................... 8.74

Jan 1, 1920