Search Documents

By Abid O. Adekunte, David Greentree, David Hilton

"The construction industry is witnessing an upsurge in demand by property developers to have basements incorporated into their residential and commercial development schemes. This is attributable to perceived improvement to aesthetic and commercial values of buildings by underground space, amongst property owners. This trend is compelling more owners of buildings with no basements to explore the possibility of creating underground additional space beneath their existing buildings. The creation of a basement beneath an existing building entails the installation of a reinforced concrete (RC) underpinning system, temporary works, bulk excavation to basement formation level and construction of the basement structure, with associated aesthetic finishes.The L-shaped segmental RC underpins are a common system of choice for such underpinning works. They are multi-functional; providing vertical support to the existing superstructure, providing lateral support to earth and groundwater, supporting temporary and permanent surcharge loads from adjacent structures and services, whilst also forming part of the basement slab. This paper focuses on the first function of the Lshaped RC underpins; capacity under vertical compressive loading from the superstructure. The most common industry approach for the prediction of vertical capacities of L-shaped underpins is the oldfashioned Meyerhof’s (1953, 1963) effective width method, which is based on the traditional bearing capacity theory, whilst accounting for correction factors for eccentricity and foundation depth.The traditional bearing capacity formula is surrounded by a significant degree of uncertainty; estimates are highly user-dependent, as capacity prediction is sensitive to the chosen angle of shearing resistance (?), while there are difficulties in accurately measuring the in-situ ??value of cohesionless soils. In addition, the effective width approach has been shown to be very conservative by many research workers. The segmental RC underpins presented in this paper have been designed with an alternative less conservative method, which involves the use of a recently proposed consistent work-hardening plasticity model calibrated with field-based tests on footings of various aspect ratios and relative embedment in sand. The paper highlights the principles behind underpinning works for urban basement construction, the design of L-shaped RC underpins with the recently proposed plasticity model and the advantages of the method over traditional bearing capacity solutions and pre-existing plasticity models."

Jan 1, 2016

By Matthew P. Wickham

The use of existing mine records in conjunction with an appropriate classification scheme can be used successfully to identify and characterize the composition of existing waste rock facilities where traditional approaches based on sampling may be impractical or costly. This paper describes an alternative approach used in an acid rock drainage investigation at a porphyry copper mine with numerous waste rock dumps of varying ages and composition. The approach integrated detailed lithologic and mineralogic information about the ore body and overburden materials with records of pit development and dispatch. A waste rock characterization matrix was developed for the site to meaningfully classify lithology and mineralogic composition of the waste rock. The approach resulted in a focused geochemical testing program and facilitated interpretation and application of the test results.

May 1, 2002

Orebody modelling and reserve estimation are now subject to sophisticated procedures, but folded stratiform and stratabound deposits present particular problems in transforming coordinates of samples and blocks into their correct relative positions prior to deformation. In this paper an approach is taken whereby a geologically interpreted reference plane is used as a control. The study area is the Rio Tinto Minera copper mine near Santiago de Compostela, Spain. The transformation procedure on the 'unfolded' state is described, its limitations assessed and advantages considered.

Jan 1, 1987

By K. Duckworth

The treatment of Turam data by conventional procedures does not permit the interpreter to make any estimate of induced current magnitudes. This problem may be overcome by using an alternative data treatment procedure. It is shown that the conventional and alternative procedures are equally satisfactory in terms of their ability to provide the location of line currents and estimates of conductor quality. A full description and evaluation of both procedures is presented in conjunction with field examples to show how the alternative procedure may assist interpretation when used to complement the conventional approach.

Jan 1, 1973

By D. Lopez, D. L. Doerr, T. J. Drake

Automated scanning electron microscopy has been widely used in recent years for gold and other precious metals characterization. However, application of this technology is resource intensive and in most instances needs specialized technical personnel to process the samples and information. Newmont Metallurgical Services (NMS) has recently acquired an ASPEX SEM. The ASPEX SEM offers some significant benefits/improvements in sample preparation and sample processing time. NMS and ASPEX have been working together to develop an approach utilizing this technology for the minerals processing industry and specifically for gold characterization.

Jan 1, 2011

By S Farrokhpay, C Akop

Copper-gold-pyrite ore types with an elevated pyrite head grade utilise a sequential flotation circuit to recover initially a copper-gold concentrate with a pyrite-gold concentrate being produced from the tailing of the initial circuit.A simpler flotation circuit with lower operating and capital costs for treatment of such copper-gold-pyrite ores with an elevated pyrite head assay was sought in the described work. After characterising the ore, the objective was to apply bulk rougher flotation of both the copper minerals and pyrite for the copper-gold-pyrite ore at natural pH. Hence, no lime or other pyrite depressants were required in bulk roughing. The bulk rougher concentrate was upgraded in a cleaner circuit to reject non-sulfide gangue and then the resulting sulfide concentrate was separated by depressing the pyrite to produce two products (copper and pyrite concentrates, both having significant gold assays). A standard test with a sequential circuit was applied to the same ore for comparison.The chalcopyrite and pyrite losses in the tailing from the rougher circuit were the same for both bulk and sequential roughing (less than six per cent). In cleaning (bulk circuit), the effective pH for pyrite depression was 11.8–11.9; copper recovery was 70–75 per cent (grade of 15–20 per cent copper) and the sulfur recovery (pyrite product) was 80–85 per cent (grade of 34–36 per cent sulfur).These results approached those for the sequential circuit where copper recovery was 80–85 per cent (grade of 15–20 per cent copper) and the sulfur recovery (pyrite product) was 83 per cent (grade of 38 per cent sulfur).The observed effects of feed sizing on the liberation of minerals in the rougher feed are discussed. Options for improving the bulk circuit results are also considered. Importantly, the implications of various aspects of these findings for the bulk circuit are discussed with an industrial focus.CITATION:Akop, C, Farrokhpay, S and Johnson, N W, 2014. An alternative circuit for chalcopyrite-pyrite ores with elevated pyrite content in Cu-Au ore treatment, in Proceedings 12th AusIMM Mill Operators’ Conference 2014 , pp 407–416 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 1, 2014

By Joe E. Bryan, John P. McDonnell

"Underground mine rib stability is an ongoing safety issue in all mining applications. Rib-related accidents present at least as many problems as roof-related ground control. Underground coal mine rib safety is continually emphasized as a primary safety concern at every mining operation. The application of rib support materials is increasing as more difficult mining conditions are experienced due to deeper cover, more variable depositional environments, thicker seam heights, and multi-seam mining effects. As the coal seams become thicker, the need for rib control increases. A variety of rib control methodologies has been documented and research continues on mine design methods and support techniques to reduce or eliminate rib-related accidents.This paper presents an alternative rib bolting technology using proven conventional rib support systems that allows mine personnel to more efficiently install rib support during development operations. This alternative rib bolting method allows for additional rib support with a single support installation.INTRODUCTIONRib control continues to garner the watchful eye of both concerned operators and government officials due to many factors: inherent safety within entries and crosscuts, actual rib stability history, safety improvement, cost control, improved mining methods, and maintaining production goals. As demonstrated through many years of actual practice, desired outcomes can be achieved by scrutiny, innovation, and combining improved practice methodologies.Rib surface control continues to be a necessary part of the production process. However, production and safety issues persist at the point of installation using current control apparatuses and methods to install rib controls. Among others, current issues of rib control are quantity of bolts required, overall time required for installation, ease of apparatus installation, flexibility of the system for appendage installation and operator exposure. A need exists to decrease the time miners spend installing rib control apparatuses, while increasing the safety of such installation processes. At present, installing rib control devices (primary rib bolts, plates, mesh, and, in most cases, secondary rib bolts and plates) is performed in a very congested area and is time-consuming process at the primary coal-winning area: the face. The MatLok system addresses these problems."

Jan 1, 2015

By Joe E. Bryan, John P. McDonnell

"Underground mine rib stability is an ongoing safety issue in all mining applications. Rib-related accidents present at least as many problems as roof-related ground control. Underground coal mine rib safety is continually emphasized as a primary safety concern at every mining operation. The application of rib support materials is increasing as more difficult mining conditions are experienced due to deeper cover, more variable depositional environments, thicker seam heights, and multi-seam mining effects. As the coal seams become thicker, the need for rib control increases. A variety of rib control methodologies has been documented and research continues on mine design methods and support techniques to reduce or eliminate rib-related accidents.This paper presents an alternative rib bolting technology using proven conventional rib support systems that allows mine personnel to more efficiently install rib support during development operations. This alternative rib bolting method allows for additional rib support with a single support installation.INTRODUCTIONRib control continues to garner the watchful eye of both concerned operators and government officials due to many factors: inherent safety within entries and crosscuts, actual rib stability history, safety improvement, cost control, improved mining methods, and maintaining production goals. As demonstrated through many years of actual practice, desired outcomes can be achieved by scrutiny, innovation, and combining improved practice methodologies.Rib surface control continues to be a necessary part of the production process. However, production and safety issues persist at the point of installation using current control apparatuses and methods to install rib controls. Among others, current issues of rib control are quantity of bolts required, overall time required for installation, ease of apparatus installation, flexibility of the system for appendage installation and operator exposure. A need exists to decrease the time miners spend installing rib control apparatuses, while increasing the safety of such installation processes. At present, installing rib control devices (primary rib bolts, plates, mesh, and, in most cases, secondary rib bolts and plates) is performed in a very congested area and is time-consuming process at the primary coal-winning area: the face. The MatLok system addresses these problems."

Jan 1, 2015

The current genesis of pisolites and channel iron deposits (CID) has been defined in a way that presents CID as unique in the world. The published material on detrital pisolites mostly details the West Australian CID; most other global occurrences of iron-rich pisolites would seem to have an in situ genesis. An alternate theory for the origin of the Hamersley CID is suggested, involving an older greensand protore derived from volcanic ash and deposited on a glaciated Hamersley surface. This paper presents the author's observations, sampling and mapping after several years study and exploration for pisolites and CID in WA and in some locations overseas. It is not what one would call scientifically complete, and possibly asks more questions than it answers, but it is presented as a point for discussion.

Jan 1, 2009

By Selvaganesh Selvaraju, Deepak Raj, Sandun N. Wimalasena

One of the world leading oil majors is carrying out a multi-billion-dollar expansion of its integrated manufacturing complex in Jurong Island, Singapore, the global oil trading and refining hub. Soil conditions at the site comprise 10-20m thick reclaimed sand, below which is 3-5m of soft marine clay underlain by stiff soils of the Jurong Formation. The conventional foundation solution would adopt driven precast concrete spun piles; however, precast spun piles have a high embodied carbon from the usage of steel and cement. In order to reduce the carbon footprint, an alternative solution comprising deep vibro techniques was developed. The EFFC-DFI carbon calculator was used to demonstrate the expected scope 1, 2 and 3 carbon emissions for both driven pile and vibro techniques. By using in-house, state-of-the-art fuel-efficient machines, alongside the innovative design concept and usage of natural materials such as sand and stones, actual CO2e emissions were reduced from a potential 21,000 tonnes, with the spun pile’s solution, to only 1,900 tonnes. The saved 20,100 tCO2e is equivalent to an average car driving 2,000x around the Earth’s equator. The paper discusses the application of soil improvement as a more sustainable, lower carbon alternative to conventional piled solutions.

May 1, 2022

By Yotamu Hara, Animesh Jha

"High Curie temperature and large magnetic anisotropy make cobalt an important magnetic material for power engineering applications of magnets operating above ambient. It is also an essential constituent of modem lithium - cobalt oxide batteries, without which modem electronic gadgets will be functionless. Both these applications of cobalt save energy and reduce C02 emission. Ironically, however the production of cobalt from copper and nickel smelting in the world remains one of the least energy efficient process. The paper analyzes the energy balance, material and process chemistry of overall cobalt recovery and the waste generated, which entail the process limitations. We discuss a novel approach for the reduction of sulphide minerals with CaS04, permitting better separation of metals from slag and generating higher concentrations of S02. Such an approach may seem to be an alternative energy efficient option for processing and further purification of cobalt in metallic form. The results are discussed in the context of process thermodynamics and reaction kinetics for Coalloy production.IntroductionCopper sulphides bearing minerals are predominantly associated with iron and cobalt sulphides minerals. Carrollite (CuC02S4) is a mineral that has both copper and cobalt sulphides [1,2], which may also contain FeS, by substituting either Cu2+ or Co2+ ions in mineral lattice. On smelting copper - cobalt - iron sulphide concentrates, some parts of the cobalt often partition to matte at preferentially low oxygen potential by forming eventually an alloy phase with copper, however a significant fraction is often lost into an oxide slag rich in iron and silica. The slag heaps accumulated in the Copperbelt region in Zambia from last 60 years of mineral processing and metal extraction is reported to be 0.3 to 2.6 wt% of cobalt element, which is high [3]. Cobalt is also available in smaller concentrations in the copper smelting and converter slag, as reported in the investigations on the recycling of such slag materials using leaching with ammonium salt [4], iron chloride [5], and sulphuric acid [6], and solvent extraction [7].The recovery of cobalt from slag, irrespective of the techniques researched so far, does not seem energy efficient because the molten slag at 1300°C is quenched by losing thermal and chemical energy. The quenched slag is again reheated to above room temperature for leaching and above liquidus for carbothermic reduction of residual cobalt oxide. For example, in Mintek DC arc smelting [3], 80% of cobalt and 95% nickel are recovered by expending 600 kW per tonne. Assuming that the mineral carrollite is the starting phase, the univariant calculations for the Co-Ca-Si-S-0, Fe-Ca-Si-S-0, and Co-C-Si-S-0 systems are shown in Figure la, lb, and le, respectively at 1573 K."

Jan 1, 2012

By David A. Neudorf

Known resources of nickel laterites considerably exceed nickel sulfide resources and there is a growing consensus that most new nickel production will be sourced from laterites. The development of nickel laterite projects is very capital intensive due to inherently low grades, inability to concentrate ore prior to processing, complex processing technology and typically large infrastructure requirements. This has resulted in most recently proposed projects being "mega-projects," to achieve economics of scale. However, this may result in unacceptable project risk in many cases. Furthermore, many known laterite resources are not large enough to support a mega-project. This paper describes an alternative model for nickel laterite project development that allows, smaller scale and reduces capital exposure. The requirements of this model in terms of process technology, infrastructure and commercial innovation are described. Keywords: Nickel, Laterite, Project, Economics

Jan 1, 2004

By D. F. Caicedo, I. A. Brum, L. A. Betancourt

Ferricyanide ()()*is one of the most stable cyanometallic complexes present in the gold mining effluents. This cyanocomplex is hard to degrade by natural attenuation and generates a negative impact on aquatic environments. Although free cyanide (CN-) can be obtained by acidifying the solution, the CN- is lethal for all forms of life. The oxidation of ()()* in a typical photocatalitic system was evaluated with the addition of H2O2. To verify the degradation, chemical parameters, such as free cyanide, the formation of ammonia, nitrate, and total iron were analysed at the end of the process. The degradations obtained were up to 70% in a heterogeneous photocatalysis system with TiO2, and a low rate of dissociation was observed by photolysis. The best result was achieved with the H2O2 and TiO2 photocatalytic system, and the stoichiometric concentration was about 2.5 times less than the peroxide used in the real gold mining industry, reaching 83% degradation. The photocatalytic process obtained less toxic byproducts than the original synthetic ferricyanide used as mining wastewater. Keywords: Cyanide, heterogeneous photocatalysis, titanium dioxide, UV radiation, gold wastewater

Jan 1, 2020

By Rimas Veitas, James Panton, Erich Steinlechner

"Ductile Iron Piles (DIPs) are a proven, viable alternative to conventional piles, and for the past five years have been successfully used throughout New England. The installation process of DIPs lends itself to both large and small projects. Because of their installation efficiency they provide a substantial cost savings over drilled micro piles. Several New England projects (both end bearing and friction) will be evaluated, focusing on the installation and design procedures.BackgroundThe driven ductile cast iron pile (DIP) is a simple, fast and highly effective pile system. Over the last 30 years over 10 million feet of this pile system has been installed worldwide predominantly in Europe. DIPs are a prefabricated driven pile system utilizing high strength ductile iron pipes which are manufactured using a spun-cast process having outside diameters of 4 5/8” and 6 5/8 inches. The standard pile lengths are 16.4 feet long. The pipes are manufactured with a tapered socket with an internal shoulder for full engagement at the top and a tapered spigot at the bottom. The individual pile sections can be connected with this Plug and Drive® connection to drive a pile of any length.(see figure 1)The DIP system’s inherent advantage is the simplicity of the installation process. The piles are driven to refusal or to a required penetration depth with the use of a high-impact highfrequency hydraulic hammer (ie. Breaker hammer). In most applications the hydraulic hammer is mounted on an appropriately sized excavator (see Photo 2). At the completion of driving the pile, the pile is cut off (see photo 5) and the remaining section of pile is fitted with the appropriate driving shoe and serves as the lead section for the next pile. Being able to reuse the cut off section, yields minimal if any pile shaft waste. The pile is then fitted with an appropriate pile cap plate to receive the load from the superstructure."

Jan 1, 2017

By R. W. O. Kersten

"The aim was to investigate stable pillar design procedures currently in use on chrome and platinum mines, subject them to a critical appraisal, and review some of the alternatives proposed in the literature. It is concluded that these pillar design methodologies suffer from drawbacks that can be detrimental to the mining industry and that an improvement is essential.It was decided that the increased availability of analytical models and failure criteria could be exploited to update the empirical to an analytical/empirical method. The proposed methodology adopted FLAC2D simulations using the Hoek-Brown failure criterion to calculate full stress-deformation curves for typical pillars. The mine stiffness concept was introduced to determine the pillar load, integrating the influence of the pillar and strata stiffness and mining geometry.The proposed methodology was calibrated by comparing experimental and predicted deformations and failure depth of pillars in one hard-rock mine. The conclusion was that the methodology is an improvement over the one currently in use. An improved factor of safety is given by the intersection point of curves for the stiffness of the system and the pillar strength. IntroductionThe paper describes part of the findings of a PhD thesis (Kersten, 2016) on an alternative stable pillar design methodology. This publication contains sufficient information to implement the method for further interpretation and testing. A detailed comparison of measured and predicted data is dealt with in the thesis, and the results show that there is promise in the methodology.The design of stable pillars in mining is of fundamental importance, not only in the Bushveld Complex mines but throughout the mining industry. Wherever mining occurs, pillars are formed at some stage and it is essential to predict their behaviour, i.e. whether they will burst, yield, or remain stable. Although of crucial importance, pillar design still suffers from major shortcomings and weaknesses."

May 1, 2019

By Joseph A. Cybulski

In response to the mining of reserves with increasingly adverse round conditions and the roof support recommendations of current rock mechanics theories, roof support systems have continued to improve. A few of these newer bolting systems. while giving improved performance, require a new method to check bolt installation. Previous1 , a torque wrench had been the sole means to check the installed torque of a roof bolt. However, due to the characteristics of soma of the newer bolting systems which can prevent a reliable manual torque check and a recent change in the federal roof control regulations, an alternative method to a manual torque check is gaining acceptance. This alternative method improves monitoring the hydraulic pressure of the roof bolting machine torque circuit and provides a safe, reliable and quick wav to check bolt installation in those instance when an accurate manual torque check cannot be made.

Jan 1, 1990

By Fourney W. L, Weaver T. A

A large number of single borehole crater-forming shots were analyzed in terms of correlating the crater parameters with explosive weight. The objective was to determine a scaling rule for which the explosive weights varies over many orders of magnitude and covers many different geologic materials. Included in the data base are type and weight of the explosive, geologic medium and degree of saturation, charge depth-of-burial, crater dimensions, and crater rim (lip) measurements. An alternative scheme to the usual plot of scaled depth-of-burial versus scaled volume removed was desired. The proposed technique uses a plot of the logarithm of the explosive weight, Y, versus the logarithm of the slant length, R', calculated from the known depth-of-burial of the explosive and the measured crater radius. A plot of Log Y in kilotons versus Log R' in meters from all the crater tests in this data base resulted in a straight line with a of slope 0.31 and showed good correlation between the variables Log Y and Log R'. Since the crater data includes many geologic materials, different explosives (including nuclear tests), and large differences in charge weight, the straight line fit suggests that the type and strength of the geologic materials play a subordinate role to explosive energy in crater formation.\

Jan 1, 1990

Recent work has shown that, in at least some instances of regional metamorphism, metamorphic diffusion has been restricted to such relatively minute distances (of the order of 1.0 mm and much less), that there is no clear, direct evidence of prograde metamorphic mineral reactions, and that metamorphic equilibrium does not appear to have been attained through even very small domains.A number of stratiform orebodies and their immediate metapelitic environments exhibit very large, seemingly "disequilibrium", assemblages of regional metamorphic index materials. One such, in north-eastern Australia, includes laumontite, prehnite, chamosite, stilpnomelane and related hydrous iron silicates, primary chlorite, sericite-muscovite, epidote, actinolite, biotite, almandine, andradite, metamorphic hornblende and clinopyroxene, scapolite, staurolite and sillimanite, all within a 20 metres length of a single diamond drill core. In this and several related occurrences it appears that metamorphism has involved very small scale in situ transformation of pre-existing materials rather than large-scale diffusion and reaction, and hence that the observed metamorphic minerals have probably been derived directly from precursor materials of sedimentary/diagenetic origin.It is concluded that the fluctuating seafloor interplay between warm, acid, concentrated hydrothermal waters with cold, alkaline, dilute seawater during stratiform ore formation, together with diagenetic processes, leads to the formation of a wide range of clay, chlorite, mixed-layer clay and clay/chlorite, and zeolite precursor materials and hence, later, to the complex assemblages of metamorphic minerals closely associated with some stratiform ores. Commonly included in highly localized assemblages of this kind are various minerals which, in conventional regional metamorphic studies, are generally regarded as indicating widely different grades of metamorphism.

Jan 1, 1982

By D J. Williams

Tailings management is typically constrained by many factors that are taken as given. These include the tailings production rate that must be accommodated, the desire to deliver the tailings cheaply and easily as a slurry using centrifugal pumps, and a tailings footprint limited by the desire to minimise capital costs leading to a rapid tailings rate of rise and a low settled density. Other constraints include wall raises being delayed due to limited capital, inadequate tailings water and rainfall run-off management, and the net present value (NPV) approach to capital expenditure that puts off expenditure at the expense of performance. All of these constraints result in the tailings occupying much more volume than is necessary, being difficult to rehabilitate, and not being capable of supporting a beneficial post-mining land use or function. Ultimately, such constrained tailings storage facilities will assume very large dimensions, but each will typically be developed as a series of smaller storages that are filled rapidly and added to progressively. The optimal stage to efficiently recover process water is in the process plant, but this is limited by the desire to use centrifugal pumps rather than the more costly positive displacement pumps, which also require greater management of tailings deposition. Process water can also be recovered from the tailings storage facility (TSF) decant. Tailings water that is not recovered from the plant or the decant will be lost to entrainment within the tailings, seepage to the foundation or through the wall, or evaporation. However, tailings water is not always suitable for reuse in the plant, particularly when it becomes contaminated by the processing, and it may simply be stored on the TSF. This will exacerbate the seepage of contaminated water to the environment, and will limit the densification of the tailings that would otherwise occur due to desiccation of the tailings on exposure. If a larger tailings footprint were developed early, cycling of tailings deposition in thin lifts between a number of cells could occur, maintaining the dormant cells dry to enhance desiccation and densification, and ultimate rehabilitation. The end result would be a much reduced stored volume of tailings, the potential for upstream raising, and ease of rehabilitation to a high level. The higher initial capital cost would be overtaken by reduced ongoing capital, operating and closure costs.CITATION:Williams, D J, ?2014. An alternative whole-of-life approach to tailings management, in Proceedings Life-of-Mine 2014 , pp 285–298 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 16, 2014

By Frederick Bacorn

THE paper of Reno H.. Sales on Ore Deposits at Butte, Mont.,' is a careful and painstaking work, an important contribution to the literature of the subject. As is almost inevitable in a work of such magnitude, a few misstatements occur; and some of these I shall endeavor to point out. Mr. Sales has also fallen into error, as I think, in the application of his theory of ore deposition; and while the modification which I venture to suggest is a minor one geologically, it is of great economic importance. I refer to Mr. Sales's idea of an outward migration of mineralizing solutions from a supposed central copper zone. In order that I may make myself plain, I must first epitomize the theory as a whole, which I understand to be this: All of the Butte ore deposits (save those along the Continental fault, not herein considered) are derived from solutions of magmatic origin, which, at the beginning of their ascent from a great depth, were, acid in reaction, hot, and were carriers of copper, zinc, manganese and other metals. As they departed from their point of origin they lost heat, and, because of reactions with the wall rock, became more and more alkaline. The metal deposited from these solutions in their hot condition was principally copper; and, as they cooled, they deposited less and less copper and more and more manganese and zinc, until the copper practically disappeared. With this theory in general I have no fault to find, nor have I, it must be stated, the qualifications necessary to enable me to pass upon it critically; but it seems to me to be reasonable and probable, and is one which I am quite willing to accept.

Jan 8, 1914