Search Documents

By Alain Simard, Sylvain Lacroix, Louis-Martin Dubé, Pierre Pilote

"The Harricana-Turgeon belt (HTB) is a poorly known volcano sedimentary belt in the NW part of the Abitibi Sub-province; it includes the Matagami, Joutel, Brouillan and Casa-Berardi mining districts. The HTB is dominated by volcanic rocks, which covers 65% of its surface area, followed by sedimentary (18%) and plutonic (17%) rocks. It is also transected by numerous deformation zones located either at the contacts of volcano-sedimentary units and granitoid bodies or cross-cutting them. A preliminary geological synthesis shows that mineral deposits are associated with certain regional geologic elements.The volcano-sedimentary units are grouped here into twelve band-shaped lithotectonic domains, most of which extend for more than 150 km E-Win the HTB. Each domain is defined by lithologies and facies associations, and gradiometric and electromagnetic signatures.Eight domains are composed of basaltic or basaltic to komatiitic associations, interlayered with thin pelagic sediments, representing sub-marine lava plains. Two domains are composed of basaltic to rhyolitic associations that are interpreted as volcanic arcs formed by one or several central volcanic complexes.The sedimentary rocks are grouped into two lithotectonic domains. Each is composed of rhythmic sequences of turbiditic sandstone- siltstone-shale, Algoma-type banded iron formations and conglomerates bearing plutonic and volcanic pebbles. These three facies come partly from mass flow at the bottom of a slope in a moderately deep marine environment. In each domain, a conglomerate sandstone- siltstone sequence, containing poly genetic pebbles of local provenance, is similar to Timiskaming-type alluvial deposits.The nineteen granitoids, found within the HTB and at its border, belong to numerous petrographic suites, and have been grouped into four structural families: pre-tectonic, pre- to early-tectonic, syn- to late-tectonic, and late- to post- tectonic. The pre- to early-tectonic plutons are presumed to be sub-volcanic and are generaly associated with the volcanism of central complexes.The HTB mineral resources consist essentially of polymetallic (Cu-Zn-Ag-Au) and gold (Au-Ag) deposits. The polymetallic deposits are located exclusively in the two lithotectonic domains composed of basaltic to rhyolitic associations (Matagami-Brouillan and Joutel Raymond). All gold deposits are associated with deformation zones superimposed on lithotectonic domain contacts and highlighted by graphitic units. Most of the deposits are hosted in the Casa-Berardi and Douay deformationzones, located at the contact of the Cartwright basaltic to komatiitic domain and the Taibi sedimentary domain"

Jan 1, 1990

By Ihsan Berk Tulu, Gaobo Zhao, Deniz Tuncay, Syd S. Peng

Longwall mine instrumentation for measuring abutment stress distribution, mining induced immediate roof stress change and deformation, and primary/secondary support loads and deformations provides critical guidance for the design of gateroad chain pillars and roof support. Field measurement results are also recognized as the best way to calibrate numerical models that have been used to study the failure and support mechanisms. International Conference on Ground Control in Mining (ICGCM) started in 1981 and since then, there were 26 papers published in the proceedings discussing the stress and load measurements as the face advanced using various field instrumentation techniques. The focus of this paper is on summarizing the analyses of the data published in these papers. Stress/load change monitoring included the following categories: gateroad chain pillar, abutment pressure, bolt load, and pre-driven recovery room. The data shows that the magnitude of stress changes as the face advances varies considerably from panel to panel and mine to mine. However, regardless of types of stress changes, pillar stress, abutment pressure, and bolt load, there exists a rough trend of stress increase as the face advances, generally consisting of three stages, i.e., stress/load (1) initially increases steady or slowly, (2) increases rapidly, and (3) increases reaching the peaks and dropping down slowly or sharply. The stress/load increases start earlier during the second panel mining and are larger than the first panel mining. This review makes it easy for the interested persons to find valuable data for other similar projects.

By C. Roos

"Nearly every input to a mine plan is based on an estimate. The estimates may be from sample data, historical information, models, or personal opinion, but in all cases, these values are simply expected values (means). In real life, we do not get to iterate the exact conditions at our mining operation many times to ensure that that average value is attained. The expected value also tells us nothing about the spread of values that that input might take on. The result is a significant amount of unquantified uncertainty in our mine plans. Unfortunately, it is often too expensive or time consuming to update planning processes and software. This paper presents a proof of concept spreadsheet scheduling tool that can be utilized to incorporate many geologic realizations (simulated models) into the mine scheduling process. While this POC is not intended to be a detailed model of uncertainty, it is to show that there are stepping stones available for mine planners to begin to embrace uncertainty and produce more achievable plans without requiring a significant change in their planning process. INTRODUCTION Mining companies have included quantifying uncertainty, or at least risk, as a requirement of project justification for many years, however many times this process lags behind the traditional mine planning process and the results are rarely used to improve the plan. Nearly every parameter used in a mine plan is an estimate and the mine planning engineer should recognize and even embrace the inherent uncertainty in those parameters and its effect on their forecast results. While many of the uncertainties are beyond the scope of the traditional mine plan (e.g. global market uncertainty, socio-political uncertainty, environmental uncertainty), some uncertainties can, and should, be included in the mine planning process. Many mining engineers have incorporated techniques such as Monte Carlo Simulation or Discrete Element Simulation to understand the range of productivity that can be expected for an operation but most still rely upon one estimated model for what is likely the critical component of technical uncertainty, the geologic resource. Modern geostatistical methods can provide multiple ore body realizations to provide a thorough unbiased “picture” of the geologic uncertainty but most traditional mine planning techniques are not formulated to allow for multiple (100-200) realizations of each material quality/grade parameter. Commercial software companies and research groups are progressing with development of tools that can incorporate geologic uncertainty, and even optimize extraction plans in the uncertain environment. Unfortunately, this author believes that a significant number of mining operations will not be able to adopt these techniques due to the computational requirements of the advanced techniques and the costs associated with implementing proprietary software. In an attempt to encourage mining engineers to embrace the inherent uncertainty in their mine plans, this paper presents a cost-effective alternative that can be developed in-house, at any mine, to schedule mid to long-term mine plans while continuously monitoring the uncertainty in the results. The author hopes that by demonstrating the usefulness of such a system that mining companies will encourage engineers, geologists, and geostatisticians to find ways to incorporate this powerful data into their processes and procedures."

Jan 1, 2017

By Hannu J. Häkkinen

Metal clusters exhibit unique size-dependent physical and chemical properties that differ from those of bulk materials or large particles. Here we discuss recent theoretical investigations of the catalytic properties of free and MgO-supported gold clusters. The catalytic activity of gold clusters depends sensitively on the environment, charge state, and chemical composition of the catalyst. Relativistic bonding effects stabilize large planar structures for gas-phase gold clusters, and non-compact ?quasiplanar? structures are present as important isomers also for surface-supported clusters. The hybridization of the atomic 5d and 6s states is significantly enhanced by the relativistic effects, and large spatial variations in the nature of the local electronic structure make the low-coordinated gold atoms preferable adsorption sites for oxygen molecules. The adsorbed oxygen molecules are activated to superoxo- and peroxo-states via charge-transfer from gold, and Eley-Rideal and Langmuir-Hinshelwood reactions with CO proceed via low-barrier formation of intermediates, which do not require O-O dissociation as precursors. This facilitates the low-temperature (below 300 K) activity of gold nanocatalysts in CO oxidation. Other gold nanostructures, such as mono-atomic gold nanowires, may also become important prototype systems in chemical sensing and nanocatalysis; here we show that they are active promoters of hydrogen dissociation.

Oct 1, 2003

By Donald S. Skufca

This chapter focuses attention on major factors pertaining to plant design and general arrangement of equipment. Considered are efficiency of operations, coordination of maintenance and repairs, plant testing of reagents and possible future plant expansions. The design and general arrangement of equipment considers accessibility for maintenance and isolation for repairs, minimum personnel for operations and cleanup, safe working conditions with minimum interruptions to operations. The constant introduction of new reagents requires eventual plant testing and the manner of isolating a “line” or part of the facilities to allow comparisons in the testing of new reagents is discussed. As technology improves and process changes occur, additional equipment or plant expansions are inevitable. Space for these eventualities must be considered while the general arrangement drawings are in the development stages.

Jan 1, 1986

By Marc T. Filigenzi

Two of the major ground control safety issues confronting underground mine operations today are shaft pillar stability and the failure of rock around active mine openings. Failure of a mine shaft can lead to the entrapment of workers. Failure of rock around active underground mine openings can lead to roof falls, which in turn can result in worker injuries and fatalities. Finite-element analysis has proven to he a reliable method for predicting stresses and displacements around underground mine openings. However, this is a complex and time consuming technique and is not used as often as it could be in mine planning. The purpose of this paper is to demonstrate one technique developed at the Spokane Research Center that allows the user to create a finite-element model of a two-dimensional section of an underground mine in a relatively straightforward manner. This model is then used to calculate stresses, displacements, and safety factors around mine openings. With this information, mine planners can evaluate the stability of mine openings as well as the stability of pillars and mine shafts. This analysis will help develop mining sequences and layouts that minimize stresses in that section of the mine. This, in turn, should minimize the occurrence of shaft failure, roof falls, and other hazards associated with underground mining.

Jan 1, 1997

By Raja V. Ramani

The potential usefulnuss of reclaimed surface-mined land has received much attention in recent times. This attention is partially due to the establishment of federal environmental reclamation standards but concern for achieving better land uses has also been shown by interested members of the mining industry. In an effort to develop a procedure for evaluating postmining land use potentials, specific mines in various regions of the U. S. were studied to determine the extent of current mined-land planning and to discover characteristics that are regional in nature. As part of this investigation, a 430-acre surface mine in Clear-field County, Pennsylvania has been studied. Established site planning principles have been applied to the case study site including formation of a data base, development of alternate land use plans, and evaluation of the economic, environmental, and social impacts of the land use alternatives. Key factors affecting site realization have been identified along with positive suggestions for applying land use planning processes to mined-land planning.

Jan 1, 1983

By Llewellyn A. Rubin

Under Bureau of Mines Contract H0122007, Telcom, Inc., with the assistance of Fenix & Scisson, Inc., conducted a 17-month research and development project aimed at provid¬ing a field-test model of a guidance system for use with the Bureau of Mines Probe Drill. The Probe Drill is a part of the Bureau's Mine Rescue System developed under Contract H0101262. Volume I of this report is subdivided into seven sections. Volume II consists of a number of supporting appendices. In addition, there were delivered under the Contract, a number of other documents which are referenced in the text. Section 1 deals with essential contract information, Program Background, and a general system description. Section 2 covers the project chronology in brief. Section 3 is a digest of design considerations and implementation related to the requirements of the Contract. Section 4 is a description of the System, including illustrations of the hardware, and software. Section 5 is devoted to Operation and Maintenance. Para. 5.1 describes step-by-step field assembly and disassembly of the downhole unit. Para. 5.2 covers Operation. Para. 5.3 covers various aspects of maintenance from the lowest field echelon possible. Higher level maintenance must be accomplished in the laboratory, for the present field-test model. Para. 5.4 covers system calibration. Para. 5.5 covers detailed step-by-step operational sequences. Section 6 covers the results of the field testing, post-test evaluation, and recommendations. Essentially, we conclude that the present hardware (field-test model) has proven every aspect of system performance, but failed to provide reliable assurance of an operationally ready system. Telcom, therefore, recommends a three-phase continuing effort devised to provide the operationally-ready model of the system.

Jan 1, 1973

By J. P. A. Hortley

In conventional electrolytic copper production, looped copper starter sheets, on which the copper is deposited, are supported by hard-drawn copper bars. These bars have cross-sectional areas of between 1 and 1.4 in2. The suspension bars serve two purposes: firstly to support the cathode, which can weigh up to 150 kg, and secondly to introduce the electric current. IYI examined these two requirements in detail to see whether a better design could be achieved. Considering firstly the current carrying requirement, it wee noted that under normal refinery conditions, each cathode carries between 400 and soak. If one applies the generally accepted rule that 1 in of copper can carry 1000A4 a suspension bar need theoretically contain only 1/2 in2 of copper. This contention is supported by the fact that the 500A current passed through the cathode also appears to pass quite satisfactorily through the two support loops, which have a cross-sectional area of approximately 0.35 in 2. Turning now to the strength requirements, the cross-section of hard-drawn copper bar required to support a 150 kg cathode is relatively small, and easily met by the present designs. It is not, however, the static load applied by the 150 kg cathode immersed in the electrolyte which causes the damage to the hard-drawn copper, or even the increased static load of a 100 kg man standing on one foot in the centre of the bar; the damage is actually caused by the dynamic loads applied to the bare when they are dropped, thrown into containers, lowered too rapidly into the cell, or otherwise mistreated. The important requirement is therefore that the bar does not permanently bend. Its proof stress of limit of proportionality, rather than the amount that it deflects under static loading, should be the strength criterion. With the hard-drawn copper normally used for the manufacture of suspension bars, it has been observed that the copper becomes annealed during use, particularly at the end of the bar where the current is introduced. Thus, after a relatively short time, the cathodes are in effect being supported by annealed copper bars.

Jan 1, 1978

By R. R. Sacrison

The Barrick Mercur Gold Mine lies in the Oquirrh Mountains of northern Utah. Limestone, sandstone, shale, and chert are the predominant rock types being mined. Recent expansion has encountered rhyolite. Lithology and structural attitudes control most mechanical behavior. Slope steepening, mine planning, and site selection are presented as means by which geological engineering has contributed to the economics of the operation. Geomechanics investments of approximately $700,000 have resulted in cost savings or additional production worth over $16,000,000.

Jan 1, 1990

By Thomas P. Mucho, Thomas M. Barczak, Dennis R. Dolinar

Maintaining ground stability in the gate roads, particularly the tailgate, has always been critical to the success of longwall mining, both in terms of safety and productivity. Several new support technologies have been developed in recent years to replace conventional wood and concrete cribbing for secondary roof support. Since their performance characteristics are unique, the best practices that have been developed with conventional wood cribbing may not be applicable for these alternative support technologies. Therefore, with so many options to consider and the importance of achieving adequate ground control at minimal cost, the trial-and-error approach to longwall gate road support is no longer prudent. This paper discusses a design methodology for standing secondary tailgate supports. This design technique requires in-mine measurements of tailgate support loading and convergence to establish a tailgate ground reaction behavior based on support and strata interaction. The methodology uses the performance characteristics generated in the National Institute of Occupational Safety and Health's (NIOSH) Mine Roof Simulator (MRS) to match the stiffness and load characteristics of various supports to the measured ground reaction behavior. It can be used to determine the appropriate application of alternative roof support systems or to design in-mine trials so that a fair and equitable comparison of different support systems can be made. A case study of the methodology at a western Pennsylvania mine site is presented in the paper, including a comparison of four alternative support technologies to the conventional wood and concrete cribbing historically used at this particular mine.

Jan 10, 2000

By Tianzu Yang, Weifeng Liu, Lin Chen, Wei Chen, DUCHAO ZHANG, HUI XIAO

Carbonaceous refractory materials are widely accepted in many reducing smelting processes such as iron and ferroalloys. Oxidizing processes such as copper concentrate smelting traditionally use magnesia and/or alumina/silica refractory materials. The main objection to the use of carbon-based refractories as the exposed working lining in oxidizing applications is the potential for free oxygen to react with the carbon at high temperature, which could rapidly degrade the refractory lining. In 2013, Outotec and GrafTech conducted joint trials of a thermally conductive, carbonaceous lining for smelting copper concentrates in the Outotec® pilot-scale top-submerged-lance (TSL) furnace in Dandenong, Australia. This testwork had two equally important goals; to demonstrate the ability of carbon refractories to withstand oxidizing process conditions, while ensuring that neither furnace operating conditions nor process stability were negatively affected by the refractory material. The paper discusses the input parameters of the trials, key observations during the testwork program, and the results and conclusions from both process and refractory design standpoints.

Jan 1, 2016

By A. J. Friedrich

Conveyor belts are designed to go unnoticed. When they work well, they attract little attention. When they are not doing the job right, it is noticeable, and they need to be fixed immediately. Vulcanising contractors compete for the maintenance of belts in mining/quarrying and general materials handling worldwide. Damage repair is approached in similar fashion. The damaged section of belting is normally cut out and discarded, and the belt re-spliced using hot vulcanising, Cold splicing, or mechanical fasteners/clip joints. However, there is a cost-effective alternative, alleviating the need to remove damaged sections which are often perfectly recoverable. This is the use of cold-curing repair resins and adhesives.

Jan 1, 2000

By D. J. Chaiko

Certain radionuclide ions (e.g., TcO4) exhibit unusually strong Affinities toward the polymer-rich phase in aqueous biphase systems generated by combinations of salt solutions with polymers such as poly(ethylene glycol) (PEG) and poly(propylene glycol) (PPG). Thus, aqueous. polymer ,phases could potentially be used to selectively extract these ions during pretreatment of radioactive tank wastes at Hanford. To help develop a fundamental understanding of the interactions between various ions and polymers in aqueous solution, interaction enthalpies between sodium -perrhenate and a random copolymer of PEG and PPG (UCON-50) were measured by microcalorimetric titration. An entropy compensation effect was observed in this system in which changes in enthalpic interactions were balanced by entropy changes such that the interaction free energy remained constant and approximately equal to zero.

Jan 1, 1999

By T Schrimpf, I Bryan

The life cycle of a mining project begins with initial regional exploration and ends with mine closure and rehabilitation. Between resource definition and the commencement of mine construction is the study phase. In this phase technical concepts and processes are defined, evaluated and assessed with the aim of establishing whether the deposit is economically feasible. The study phase usually involves various steps from a concept/scoping study to prefeasibility and ultimately definitive feasibility study. The complete study phase can take several years. Although the costs for studies are low in comparison to actual construction and operating costs of a mining project, they are significant especially for junior exploration and mining companies advancing projects to the point of being funded.The studies aim to determine selected parameters for the project, eg mining and processing methods, operating strategy, etc and provide a roadmap for project development ahead. Although mining studies are of the utmost importance for the mining project and the project owner, they are usually carried out using a conventional engineering approach: the consultant or design engineer works to a defined scope of work. The resulting design might be based on good engineering, but might often also fail to provide the solution for the project owner that delivers the best shareholder value. An alternative to this more traditional approach to project evaluation is what we have termed a value engineering approach. Value engineering is a creative, organised effort, which in the case of a mining study analyses the requirements of the study for the purpose of achieving the essential outcomes with the highest shareholder value. The outcome of a study is a concept that delivers the highest achievable value of the resource project and shows the way ahead to realise this value. Through a group investigation by experienced, multi-disciplinary teams the value of the resource is improved through the study of alternative options, methods and processes. The scope of the study is not limited to a narrowly defined engineering scope, but to the creation of value.The study commences with the identification of key value drivers and risks, based on the available information (drill data, geotechnical data, metallurgical test work, etc). This enables the quantification of key uncertainties, risks and value adding opportunities. Study efforts concentrate on key value drivers and are not unnecessarily spent on side issues. This approach can significantly reduce the time frame and costs for the study phase.This paper describes in more detail RMDSTEMÆs value engineering approach to mining studies and illustrates it with examples and case studies.CITATION:Schrimpf, T and Bryan, I, 2012. The application of value engineering principles to mining studies, in Proceedings Project Evaluation 2012 , pp 163-170 (The Australasian Institute of Mining and Metallurgy: Melbourne).

May 24, 2012

By A I M Ritchie

The pyrite oxidation rate is a process central to both the performance of bio-oxidation heaps, used in the pre-treatment of some refractory gold ores, and to the levels of pollution in acid rock drainage (ARD) from pyritic waste rock dumps. A goal in bio-oxidation heaps is to achieve high oxidation rates in the heap as a whole. In comparison, one way to mitigate the environment impact of ARD, is to reduce the overall oxidation rate in the waste rock dump since pyrite oxidation is the primary pollutant generation process. In both bio-oxidation heaps and waste rock dumps the overall oxidation rate of the pyritic material (the intrinsic oxidation rate) at these sites. Much research has been conducted over the last 40 years or so on the oxidation rate of pyritic material with particular emphasis on the catalytic role played by bacteria. More recently mathematical models and field measurements have clarified the extent to which oxygen transport rates are rate limiting, what the dominant oxygen transport mechanisms are and the physical properties of the heaps which have most impact on these transport rates and on the overall oxidation rate. How the performance of bio-oxidation heaps can be optimised and how the pollution load in drainage from waste rock dumps depends on gas transport mechanisms is briefly reviewed. The sensitive of the overall oxidation rate in bio-oxidation heaps and waste rock dumps to details of the intrinsic oxidation rate is quantified.

Jan 1, 1994

By G. F. Revey

All underground operations have limiting factors that influence overall profitability. Hoisting, hauling, available headings, labor, milling capacity and other factors limit how much material can be removed from an underground mine. Consequently, when mines use limited labor to handle the additional rock from overbreak and to perform extra ground support tasks, productive activity is replaced by non-productive work. Mining economics are further penalized when overbreak causes dilution of ore grades. This paper reviews organizational systems that mines have used to create environments which foster the creation of successful overbreak control programs. Actual case histories document the various quality improvement techniques, worker incentives, and practical project management systems used by these mines. A brief review of controlled blasting technology sources and overbreak measurement tools is also included.

Jan 1, 1997

By R. J. T. Charles

Introduction An apparatus that will measure the surface area of a few of the simple minerals with a high degree of accuracy has been described by J. W. Bell (1). His apparatus consisted essentially of a photometric device that would measure the turbidity of fine particle suspensions. Turbidity, in this case, was considered as a function of surface area. Bell's method of surface area determination seemed to offer possibilities of extension and improvement to the point where it might approach a general method of particle size measurement. Theoretical and Practical Considerations of the Photometric Method In order to illustrate how turbidity may be considered as a function of average particle size, and to indicate the type of results expected from a photometric method of particle size measurement involving turbidity, the following theory may be hypothecated. If a number of particles, considered as spheres of uniform diameter, are placed at random in the path of a beam of light, the following expression, which will give the change in light intensity as a function of the diameter of the particles, the specific gravity of the material, and the weight involved, can be formulated from the Beer-Lambert Law:

Jan 1, 1951

By George A. Moore

A recent survey of existing analytical results, and an attempt to correlate them with each other and with the known history and behavior of the samples, indicated that none of the commonly applied methods could be considered established as intrinsically reliable. It was. therefore, desired to attempt to establish the reliahility of some one method by external evidence. If this were to be accomplished, other methods could then be evaluated in terms of the method establibhed. The correct method was accordingly to be chosen without regard for cost, speed, or convenience; with the requirements for routine application to be reserved for 1ater consideration. The method of total combustion was chosen and submitted to a limited number of tests against primary standard samples. Considerable experimental difficulty was encountered, indicating that additional work on the experimental arrangements is desirable. While the work to (late by no means represents a completed investigation, it is helieved that the combustion method is basically sound and that its development should be continued. Choice of Method Of the several available methods, mechanical disintegration. electrochemical methods'2'3 modifications of the Sieverts experiment,% and vacuum fusion, were all judged to contain one or more factors of un- certainty not at present completely cvaluated or suhject to complete control. These methods, therefore, were considered currently unsuitable for referee purposes. Solid evolution was considered to be of doubtful significance when conducted in the alpha range. A reasonable chance for a true analysis was found for evolution in the gamma range. Preliminary tests have indicated a very slow rate for evolutioll from gamma iron. indicating a running time of possibly 100 hr. for a one-centimeter sample. Difliculty in distinguishing such slow. evolution from blank gas was anticipated, hence the method was not consitlerctl the most suitable for the present purpose. The combustion method was evaluated as being most sound on fundamental principles. The only important unknown would be to determine that the iron oxide formed did not occlude water at the tcmperaturv of the combustion.with the odvious precautions of using a sound and clean specimen, providing a system sufficiently tight to give acceptable blanks in longtime runs, and providing a suficicntly sensitive measuring system, reasonable, precise, and truthful analytical results could be logically expected. Apparatus The equipment as originally operated is shown in Fig. I, and the glassw-are in its latest modification in Fig. 2. The sulphuric acid drier and small safety tube in the upper right of Fig. I were replaced by a KOH drier and combined manometer safety tube, to permit operation at reduced

Jan 1, 1945

We have received from L. S. Cates, the following translation, by S. W. Mudd, of early mining methods in India: In Herodotus one finds a description how Darius by aid of his good horse and his good groom Oebares got himself the Kingdom of the Per-sians. He then set up in Persia twenty satrapies assigning to each a governor and fixing the tribute which was to be paid to him by the several nations. The following description of the method of getting the gold demanded of the Indians as tribute is copied directly from Book III Thalia, paragraphs 94 to 105 omitting the diversions of Herodotus which in one case consisted of a lapse of two pages following a semicolon before the author returned to the subject. The Indians, who are more numerous than any other nation with which we are acquainted paid a tribute exceeding that off every other people, to wit, three hundred and sixty talents of gold-dust. The way in which the Indians get the plentiful supply of gold, which enables then to furnish year by year so vast an amount of gold-dust to the King, is the following: Eastward of India lies a tract which is entirely sand. Here, in this desert, there live amid the sand great ants, in size somewhat less than clogs, but bigger than foxes. The Persian king has a number of them, which have been caught by the hunters in the land whereof we are speaking. These ants make their dwellings under ground, and like the Greek ants, which they very much resemble in shape, throw up sand-heaps as they burrow. Now the sand which they throw up is full of gold. The Indians, when they go into the desert to collect this sand, take three camels and harness them together, a female in the middle and a male on either side, in a leading-rein. The rider sits on the female, and they are particular to choose for the purpose one that has just dropped . her young; for their female camels can run as fast as horses, while they bear burdens very much better.

Jan 12, 1919