Search Documents

By Alden Greninger

IN a recent study1 of the deformation of metastable beta copper-zinc and beta copper-tin crystals, it was established that the parallel markings that appear on the surface of these crystals after slight deformation represent lattice transformations. Stereographic analysis showed that in beta copper-tin, the poles of these markings deviated one or two degrees from poles of the form { 133 } . In beta copper-zinc the plane delineated by the markings corresponded to no one crystallographic plane of the matrix lattice; the marking poles deviated from 5° to 11° from poles of the form {110}, and were displaced only a few degrees from (110) zones. Inasmuch as the new transition lattices produced as the result of the deformation are different for each alloy, it was thought at first that the habits of the markings might be the result of a cooperation between matrix lattice and transformed lattice. (Both matrix lattices are body-centered cubic.) Such a cooperative effect is thought to obtain in the formation of Widmanstätten figures, for it is well known that the matrix lattice alone does not determine the habit assumed by a segregate phase. On the other hand, it is not inconceivable that the dissimilarity in mark-ing habit might be due to differences in deformation characteristics of beta brass and beta bronze. Barrett and co-workers2 have shown that slip-line habits in pure iron are markedly different from those in high-silicon ferrite; and although it has been repeatedly demonstrated that plates of a segregate phase do not necessarily form on the slip planes of the matrix lattice, deformation characteristics may be more intimately allied with plate formation when a transformation is made to occur as a result of plastic deformation. However, there is the fact that the orienta-tion of markings in metastable beta copper-zinc is the same whether the markings be produced by quenching, deformation, or liquid-air cooling; likewise, quenching and deformation markings in metastable beta copper-tin are oriented in the same way with respect to the beta matrix.

Jan 1, 1938

By P. K. Naik, B. Das

Flotation studies were carried out to recover carbon values from blast furnace flue dust containing ~30.34% carbon, 4.6% volatile matter (VM), and 65.03% ash. The variables chosen for the experiments were dispersant (sodium hexametaphosphate), pH, collector (diesel oil), and frother (MIBC). The experiments were designed according to a 24 full factorial design method and regression equations for the recovery of carbon and product grade were determined from the data. Experiments were performed to optimize recovery and grade using the steepest ascent method. The maximum recovery of carbon was 90.0% with a concentration of 56.5%, whereas the maximum obtainable grade of carbon was 66.35% with 50.45% recovery.

Jan 1, 2005

By John D. Smith, Robert J. Mitchell

"There is a world-wide trend toward underground bulk mining methods employing hydraulically placed backfill. This trend is recent and therefore there is little in the literature to assist an operator in the design of backfill properties, the design of drainage facilities, the calculation of the operating forces on a bulkhead, and the control procedures considered necessary before, during and after pouring. This paper presents recommended procedures that are based on over to years of experience in the various aspects of backfilling large openings. Free-draining timber or concrete block bulkheads are shown to be a viable alternative to massive reinforced concrete bulkheads for the containment of backfill pours. Hydrostatic pressures cannot develop in backfills provided adquate drainage facilities are installed.IntroductionThe increasingly adverse relationship between mining costs and mineral values has led mining companies to develop more productive, lower-cost, bulk mining methods. This has led to the increasing use of large, more or less continuous, hydraulically placed backfill pours to enable the recovery of a higher proportion of pillar ore. The success of any bulk mining method depends on the degree of stability maintained during the stoping operations with reference to the mining rate, the performance of the backfill with respect to the per cent extraction of pillar ore, and the control of the mine structure with respect to safety and productivity. One such method has been previously described(I) where a steeply dipping orebody is being mined with a sub-level method that employs large, hydraulically placed tailings pours for delayed pillar recovery. The paper describes the rock mechanics analysis undertaken to establish the stope and pillar geometry and the approach taken to establish backfill strength requirements to prevent dilution during pillar recovery. Production with this method started in 1970 and because few, if any, large bulk pours had previously been used in- this way (judging by the lack of information in the literature), design data and criteria were needed for several aspects of the work. Methods to establish backfill quantities and pertinent ""in-place"" fill properties are described in 2, 3 and 4 in the list of references. Results of field measurements taken to establish fill pressures for bulkhead design purposes are described in 5 and 6. Laboratory large-scale model tests undertaken to confirm a fill strength design equation, which takes into account fill arching, have recently been completed. A paper describing these tests and the results will be forthcoming in the near future."

Jan 1, 1982

By B. E. Mix, L. M. Barker

THe present methods of unloading ore and coarse-crushing at the Magna plant of the Utah Copper Co. are the developments of the pas five years. Hand dumping and breaking have given way to the rotary car dump and large gyratory; and the bucket elevators and stationary screens have been replaced by conveyors and shaking screens. The coarse-crushing plant may be broadly divided into two units the primary crushing and the secondary crushing, or primary roll, unit The former was erected in 1923; with the exception of the excavation the construction was carried on by forces of the company. Excavation was started in the middle of March, the first concrete was poured July 16, erection of the steel building started Oct. 30, and the unit was com pleted and operating Dec. 30. The unit has a capacity of over 24,001 tons of ore per day, and cost more than a million dollars. Remodeling of the primary-roll unit extended over a period of years preceding this some of the work having been done as early as 1918. Delivery of Ore from Mine, Yard Storage, and Handling Ori Cars to and from Primary Crushing Plant The ore is transported, in steel cars of 80 tons capacity, from the min' to the mill, where a reserve tonnage of loaded cars is maintained. Malle locomotives, handling trains of 45 cars, or approximately 3600 tons o ore per train, deliver the trains to a storage yard which contains thre tracks—a passing track and two storage tracks. All of this trackage i equipped with an overhead trolley system, so that electric locomotive can be used in yard switching. The storage tracks have a total length of 4500 ft. and provide storag space for 140 cars. The passing track, besides permitting switchin room for locomotives, is used for storage purposes between train arrivals thus giving intermittently a total storage capacity for the yard of approxi mately 200 cars. The amount of ore stored at any one time, howeve

Jan 1, 1926

By R Wynn, T J. Akroyd, C Costa

In an environment of lower zinc prices, falling revenues and reducing zinc feed grades, Century mine implemented a fast-tracked capital upgrade program in less than 15 months. The program was designed to increase ore milling throughput capacity by 20 per cent and optimise the capacity of the flotation plant to increase zinc recovery by approximately 2.5 per cent whilst maintaining final zinc concentrate grade.A key aspect of this program was minimising capital expenditure by better utilising existing capacity and only adding capacity where absolutely required to address critical processing bottlenecks. By gaining an understanding of plant process bottlenecks and future ore blends, and by carefully selecting appropriate technologies for particular specific flotation applications, maximum return on minimal capital investment was achieved.The close collaborative work achieved between the metallurgy processing department and other operational departments on-site was also critical to the success of this project. This work increased the mining and processing capacity of the mine and port site respectively, such that the full value of the process plant modification could be achieved.Capital payback on this investment was achieved in approximately three and a half months.This case study demonstrates the extra value to a mining business that can be achieved through fully understanding the mining/processing value chain and by careful targeted action to address critical bottlenecks in the processing chain from the mine through to the port site.CITATION:Akroyd, T J, Wynn, R and Costa, C, 2014. MMG Century Mine, a case study – maximising project net present value through targeted capital expenditure and process optimisation against a backdrop of falling metal prices and feed grades, in Proceedings 12th AusIMM Mill Operators’ Conference 2014 , pp 417–426 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 1, 2014

By Louis L. Stott

It is well known that the oxide of beryllium was identified as a new "earth" in 17971 and the metal first isolated in 18272. The history of the many difficulties encountered by early investigators, the laboratory achievements of Lebeau3, Fichter4 and Oesterheld5, and the eventual development of methods for commercial production are well covered in the literature, particularly in the authoritative published work of Siemens and Halskes reviewing their comprehensive investigation of the element. Accordingly it has seemed desirable to restrict this paper to the properties and uses of beryllium and its alloys. PRopeRties of Beryllium A comprehensive summary of the available data on the properties of beryllium is given in " Gmelius Handbuch der Anorganischen Chemie7." The metal is steel gray in color. A fractured surface exhibits a bright luster, which does not appear to change with long exposure to air, probably because of a thin, transparent protective oxide coating. The specific gravity is approximately 1.8. Although there is no recorded actual determination of Young's modulus of elasticity, it has been estimateds, according to Fessenden's law, to be 30,000 kg. per sq. mm., or about 30 per cent higher than steel. Beryllium appears to have an electrical conductivity approximately 1/12 that of copper. The physical constants as available in the literature are given in Table 1. One of the outstanding properties of beryllium is its great affinity for oxygen, nitrogen, sulfur and carbon. One gram of beryllium takes up 1.774 grams of oxygen and will liberate approximately 15.23 kilogram-calories in the combination. The metal apparently reacts with all known oxides at high temperatures; in fact, beryllium will alloy with thorium and zirconium if melted in crucibles of their oxides. The true physical properties of beryllium are still somewhat in doubt, since very small amounts of impurities appear to cause a large variation

Jan 1, 1936

By Tim Siegel, III Fondren, Billy Camp

"INTRODUCTIONThe new $650 million power block at Santee Cooper’s Cross Generating Station in Cross, South Carolina is being constructed on driven pre-stressed concrete (PSC) piles. S&ME, as geotechnical subconsultant to Parsons Energy and Chemical Group, was responsible for pile analysis and design to meet the site’s geotechnical and seismic challenges. Selection, analysis and design of the PSC piles included consideration of the following:• Column and equipment support loads ranging up to 500 to 1000 kips (2224 to 4448 kN).• Incorporation of the 2000 International Building Code seismic provisions.• A thick upper soft clay and loose, liquefiable sand.• Potential for karst-related features in the calcareous Santee Limestone.GEOTECHNICAL CHARACTERIZATIONThe near-surface site geology consists of (in order of increasing depth): (1) Pleistocene sediments, (2) the Santee Limestone (considered part of the Orangeburg Group) of Eocene age, and (3) the Black Mingo Formation of the Eocene/Paleocene age (refer to the shear-wave profile in Figure 4). Pleistocene clays and silts are generally soft and will settle significantly if subjected to loading by heavy grade-supported structures while the interbedded sands are generally loose. Where submerged below the shallow groundwater table, the sands were determined to be liquefiable during a strong earthquake.The Santee Limestone is a nearly white to yellow fossiliferous and partly glauconitic rock consisting chiefly of calcium carbonate. Observations of the rock core samples from the site indicate that it can range from hard crystalline limestone (primarily in thin layers) to a lightly cemented sand. In terms of rock strength, the Santee Limestone is generally very soft to soft, depending on the cementation and other depositional characteristics."

Jan 1, 2004

By D. R. Dolinar

Geogrids can be used as surface control on both the roof and rib in underground mines to prevent the fall of small pieces of rock and material from between the primary supports. The use of screen material such as geogrid can significantly reduce the number of ground fall injuries caused by these small rock falls. Because of the potential for reducing injuries through the use of screen, the National Institute for Occupational Safety and Health (NIOSH) has conducted tests on screen materials including geogrid to determine their support characteristics and identify parameters that can affect their performance with respect to maintaining broken or failed rock. A specially designed test frame installed in the NIOSH Mine Roof Simulator (MRS) has been used to evaluate the load and stiffness characteristics of the screen material. Two types of geogrids with various designs have been tested-a polyester yarn and a polypropylene grid. The load-displacement characteristics of these grids were determined and the load capacity and stiffness were calculated through 20in of displacement. The load capacity and stiffness of the geogrid in general are shown to be related to the ultimate tensile strength of the screen material.

Feb 27, 2013

By L. J. Gurevich

IN the course of the examination, at the Bureau of Standards, of fusible tin boiler plugs for the Steamboat Inspection Service, it became evident that an investigation should be undertaken to determine the precautions necessary in the manufacture of acceptable plugs. The results of the tests of these plugs show that their manufacture has improved, though at present not all of the plugs tested are satisfactory. Table 1 gives the result of tests of some 1500 plugs from about 100 makers. Records of the tests show that several of the makers were able to produce satisfactory plugs with every pouring. The Steamboat Inspection Service has the following specifications for fusible plugs: "Fusible plugs for use in boilers of steam vessels under the Jurisdiction of the Steamboat Inspection Service shall be made of bronze casing with the bore tapering continuously and openly from end to end, and filled from end to end with tin not less than 99.7 per cent. pure and to contain not more than 0.1 per cent. of lead and not more than 0.1 per cent. of zinc. The small end of the bore may be countersunk not more than 1/32 in. in depth and width, but no recess, thread, or cavity other than this countersink shall be allowed."

Jan 8, 1919

By A. U. Seybolt

The pressure-temperature relations required to maintain the existence of Si3N4 in silicon-iron of 0.44-11.7 pet Si have been partially explored. The critical data obtained have allowed the calculation of some thermodynamic quantities, and have shown evidence of an order-disorder two-phase field extending at 800°C from about l pet to 5.5 pet Si. THE presence of silicon nitride in electrical sheet, nominally about 3 1/4 pet silicon, balance iron, has been described by several authors'" but there has been some discrepancy among them regarding the identity of the nitride. Also, little has been reported on the quantitative thermodynamic prop- erties. In the work to be reported here, the nitride has been found to be more stable than described by Corney and Turkdogan." Several silicon nitrides have been described in the past, including SiN, Si3N4 and Si2N4 (see, for example, Abegg6 and Pascal7). However, the literature on nitrides other than Si,N, is all very old. No recent investigators using the more powerful tools available today have reported the presence of any nitride other than S3,N4. W. B. Hicke and L. R. Brantley8 in 1930 determined the dissociation pressure of Si3N4 which they made by reacting nitrogen

Jan 1, 1959

By J. H. Gwynne

[ ]

Jan 1, 1962

By Sérgio Roberto Lopes Moreira, Leonardo José Silva de Oliveira, Rodrigo Seara Martins, Rodrigo Madrona Dias, Fabiano Abrahão Moreira

O nível de inclusões presentes no aço líquido interfere diretamente na qualidade das placas produzidas por lingotamento contínuo e são determinantes na formação de defeitos superficiais em produtos laminados a frio. A reoxidação do aço, bem como a passagem de escória do distribuidor para o molde, a interação do aço com o pó de cobertura do distribuidor e com o pó fluxante, têm forte influência na limpidez das placas originadas no processo de solidificação no molde. No presente trabalho, avaliaram-se as inclusões formadas durante o início do processo de lingotamento contínuo para aços de ultra baixo teor de carbono, sob três diferentes condições de adição do pó de cobertura do distribuidor, além da própria condição padrão. O objetivo do estudo é reduzir a quantidade das inclusões e consequentemente, os defeitos superficiais em placas iniciais, por meio da relação com o teor de oxigênio no veio da máquina de lingotamento contínuo. Tal investigação visou fornecer informações para alterações nas condições do processo para melhoria da qualidade do produto final.

Oct 1, 2019

By P. A. Riveros, W. D. Gould, R. Molnar, D. W. Koren, P. Bédard

This paper reports the results of the field demonstration campaign of a new process for removing and destroying ammonia from mining/ metallurgical effluent solutions. The 50-day campaign was carried out at a uranium plant in Saskatchewan, Canada. The process was developed by adapting and integrating two well established technologies: zeolite adsorption/ elution to extract the ammonia from the effluent stream and to produce a concentrated ammonia eluate; and conversion of the ammonia in the eluate to innocuous nitrogen by applying bacterial nitrification-denitrification (NH3 = > NO3 - => N2). The data show that reasonable target criteria were met, under very challenging operating conditions. Problem areas and possible causes for poor performance are identified. Preliminary and generic cost estimates of the process are also presented.

Jan 1, 2003

By Alain Dorval

"''Purification of marine sands from Les Iles-de-la-Madeleine''This paper presents the results of the work that has been done at CRM pilot plant for «Les Silices du Golfe Inc.». Essentially the purpose of this piloting was to demonstrate, on a semi-industrial scale, the process of purification of marine sands from Les Iles-de-la-Madeleine; that had been developed by CRM Process Development Service. This purification is necessary to produce silica suitable for the glass sand and the foundry sand markets.Introduction«Les Silices du Golfe Inc.» is a company who intends to dredge marine sands around the «Les Iles-de-la-Madeleine» coasts. The material will be transported to Gaspesia where electricity, water supply, and a convenient site is available to treat it. A high purity silica intended for glass sand and foundry sand markets will be produced. The objectives of this purification process are to remove organics, calcium, feldspars and iron.Material ReceivedCRM received 50 tons of sand with Si02 varying between 80,6 and 99,3 % (see Figure.1). The highest variations in impurities were the calcium and lost on ignition. We notice direct relation between% SiO2, % CaO and % L.O.I. as you can see on the following Figures 2, 3 and 4."

Jan 1, 1994

By Keli Vanessa Salvador Damin

A produtividade de uma linha de moldagem numa fundição é determinada pela vida de banca dos moldes. No processo com resina fenólica alcalina (Alphaset), a cura ocorre devido à reação de polimerização entre a resina e o catalisador orgânico ester. O tempo no qual a reação de polimerização ocorre é influenciado por dois fatores: temperatura e a relação entre a quantidade de resina e catalisador (vazão mássica). Esse estudo procura estabelecer um entendimento entre estes dois fatores no tempo de banca, estabelecendo o mais predominante, para que assim possa-se aperfeiçoar o processo utilizando a quantidade adequada de resina e catalisador, sem perder a produtividade. Para avaliar o efeito da temperatura no tempo de banca foi realizado o teste de gel time com a resina em diversas temperaturas. Para avaliar a variação da vazão mássica da resina devido as diferentes condições de temperaturas foi realizado o ensaio com o Copo Ford. Observou-se que o tempo de polimerização da resina reduz exponencialmente com o aumento da temperatura da resina, sendo o inverso verdadeiro também. O teste com o Copo Ford mostrou que a vazão da resina é muito sensível a temperatura, aumentando quase 40% num acréscimo de 10oC na temperatura.

Aug 16, 2017

By Donald R. Rathburn, Jung K. Mok

An historical review and a survey of current practice in the application of fine dry grinding in the iron ore industry are given. Both closed and open-circuit applications are reviewed, and the advantages, disadvantages, and process considerations for these two systems are discussed. The ground ore temperature rise, and the cool-ing of ground ore and mill ventilation are also discussed, along with the agglomerating characteristics of dry ground ore. Screening during the production of high-grade lump ore for blast-furnace burden, particularly from recently developed ore deposits in Australia, South America, and Africa, has resulted in large tonnages of fine iron ore. Certain owners have decided to stockpile these fines until some later date. Others have either sold these fines "as is" for agglomeration at the steel plant or have established their own pelletizing plants to produce a premium quality agglomerate. In almost all cases some grinding is required prior to pelletizing. Most of these ores occur in the weathered state which may result in difficulty in the thickening and/or filtering stages when ground in a wet system. The resulting filter cake would probably require partial drying to provide a product suitable for the subsequent balling operation; however the partial drying and subsequent material-handling steps often produce random agglomeration. Normally, these agglomerates will cause problems in the green ball formation process. Thus dry grinding has developed as a solution to this problem. Most of the dry iron ore grinding installations utilizing the standard-type rod or ball mill are grinding high-grade natural ores that do not require beneficiation. Although power consumed per ton of final product is 120% or greater in dry grinding as compared with wet grinding, the additional power required for dewatering and handling a wet ground product can easily exceed the initial power savings. Dry grinding of raw feed and clinker has been practiced in the cement industry for many decades but its major application to the iron ore industry has taken place only within the past ten years. The successful application of dry grinding to iron ore has required the solution of several unique design and operating problems. The purpose of this paper is to review recent developments in fine dry grinding of iron ore for pelletizing and to discuss the various designs and operating parameters involved.

Jan 1, 1971

By M. A. Neri

A lead-free solder alloy Sn-57.5Bi-0.5Ag has been developed by mechanical alloying and foundry processes, and has great potential as a lead-free solder system, to low temperature applications. The mechanical alloying powders show a microstructure different to the microstructure obtained by the foundry process for the same alloy. Its melting point of 142°C is slightly higher than that of eutectic tin- bismuth solder. Examination of the microstructure revealed that eutectic microstructure is refined by the mechanical alloying process Ag additions. The lamellar structure of tin-bismuth gradually disappeared with the process of mechanical alloying, while intermetallic compounds AgaSn phase was formed. The Cu-Sn intermetallic compound layer formed at solder-copper interface is the Cu6Sn5 phase. Keywords: Lead-free solder, Microstructure. Mechanical alloying

Jan 1, 2003

By D Monaghan

Pumping water for dewatering and water supply is a major component of electricity consumption at mining operations and hence is a significant contributor of greenhouse gas emissions. Pump systems at a typical mine site can be broadly split between above ground centrifugal pumps (generally used for water transfer) and submersible borehole pumps (generally for dewatering and/or water supply). In contrast to above ground pumps, submersible borehole pumps can be subjected to varying duty points (total pumping head and flow requirements) as a result of changing bore water levels and yields, particularly in mine dewatering bores. Typically, much higher pumping rates are required at the commencement of dewatering to remove groundwater storage and to intercept sufficient groundwater throughflow to achieve target drawdowns. Once target drawdown levels are reached it often only requires much lower pumping rates to maintain target drawdowns. Pumps sized to remove the large initial volumes of stored groundwater often end up too large for the lower flow rate required for æmaintenance pumpingÆ. This often results in pumps operating inefficiently, a problem which can be made worse when pumps are throttled to reduce the pump output to better match the reduced bore yield. Current standard operating practice places little emphasis on the electricity consumed by pumps, with the focus often being on minimising capital expenditure (æIf it ainÆt broke donÆt fix itÆ). Submersible pumps can have extended service lives of ten to 20+ years. However as the capital cost of a pump typically represents only five per cent of the life cycle cost, the vast majority of the life cycle cost is energy usage.Through the implementation of a regular review of the operation of borehole pumps, significant reductions in electricity use and associated greenhouse gas emissions can be achieved without impacting the dewatering schedule. In most cases, the capital cost of replacing pumps is more than offset by savings in operating costs. These potential savings become even more significant when considering proposed future carbon tax levees.This paper presents examples of the opportunities for significant reductions in energy, greenhouse gas emissions and overall costs that can be achieved for a range of dewatering scenarios.

Jan 1, 2009

By Jian Li, J. Miles, P. Liu, W. Cook, W. Zheng, Alvin Qian

"High temperature corrosion resistance is one of the key challenges in water-cooled nuclear reactor material development. Iron based materials with low Cr content exhibits superior high temperature mechanical properties. However, due to the relatively low Cr content, their corrosion resistances under supercritical water (SCW) conditions are generally poor. High-Cr alloys, despite their much improved corrosion resistance, often suffer from embitterment due to the precipitation of sigma phase in the microstructure. In this study, corrosion resistance of an in-house produced high Cr steel (25%Cr) showed much better corrosion resistance compared to that of low Cr alloy (14%Cr). This high-Cr steel can potentially be used as a corrosion-resistant coating for structural materials in sew applications.INTRODUCTIONThe supercritical water-cooled reactor (SCWR) concept is one of the promising designs for the next generation nuclear reactor, in which the water coolant is designed to be at about 625°C and 25 MPa at the outlet of the reactor core. Under these relatively high-temperature and high-pressure conditions, the cooling water is at supercritical state. High-temperature mechanical properties, corrosion resistance and radiation damage are some of the key challenges. At present, no single commercial material can meet all the requirements of the severe SCWR in-core conditions. Canada supports research in several key areas of SCWR development, including materials selection. A major focus is on materials development, selection and evaluation for both in-core and out-of-core components.Allen et. al. [1] carried out a comprehensive study on the SCW corrosion resistance up to 600 °C on commercially available high temperature materials including Ni-based super alloys, various types of stainless steels and ferritic/matensitic steels. They concluded that ferritic steels develop a relatively thick but mechanically stable oxide layer, while austenitic alloys develop a thinner oxide layer that has been found to spall in some cases. Of the three types of alloys examined, the Ni-based alloys exhibited the lowest oxide thickness, however pitting was observed in the vicinity of intermetallic precipitates."

Jan 1, 2011

By William P. Blake

In my short " Contribution to the Early History of the Industry of Phosphate of Lime in the United States,"* mention is made of the erection by Dr. Emmons and myself of a magnetic machine for the removal of iron-ore from the so-called "red sand," a granular form of apatite. At the time of writing, it was not possible to refer to my diary and notes, and the general statements were made from recollection. The subject seems to have special interest at this time, by reason of the many forms of machines now being introduced and patented for similar purposes; and, having received letters of in quiry for further details, I have availed myself of the opportunity, when recently at home for a few days, to refer to the original notes and records made by me in 1852, when living at Crown Point and Moriah, on Lake Champlain. The machine mentioned in my former communication mas pur chased of Ransom Cook, of Vermont. It arrived, and was un loaded from the boat and taken part way up to the nine, on October 18,1852. The weight was 1400 pounds and the cost $1200. It was constructed, not with permanent magnets but with electro-mag nets. It was set up and first turned by hand on October 28th, using twelve battery-cups or cells, the connections being made with wires in small cups of quicksilver. The magnets were not sufficiently strong, and the machine did not work well. More batteries were sent for, arid the machine was started with steam-power on Novem ber 6th. The ore was dried in kilns, then put through a coffee-mill crusher or grinder, and then fed upon the leather belt which carried it under the magnets. Many experimental runs were made, and some diffi culties were overcome, before December 21st, when the record states that the machine was working pretty well, using 24 cells and giving a product of from 3 to 5 tons a day of granular phosphate. The ori-

Jan 1, 1893