Search Documents

By Yongsik Jeong, Kwangmin Kim

Most tailings storage facility (TSF) failures occur in active, upstream-type TSFs, often due to extreme and unexpected weather conditions—specifically, heavy rainfall and strong winds. This case study of a TSF in southern Arizona examines the use of the optimum safe beach distance (Doptimal, the optimum distance between the crest and decant pond) as a tool for monitoring geotechnical stability and reducing catastrophic failures. A digital model of the TSF was used to determine the critical beach distance (Dcritical) under normal weather conditions, and Doptimal was then determined by adding incremental distances to account for the effects of heavy rainfall (Drainfall) and strong winds (Dwind). To overcome accessibility issues, a drone was used to map the TSF geometry and create a digital model. This model served as the basis for conducting a coupled stress-seepage, finite-element (FEM) analysis with a safety factor of 2.0 to assess geotechnical stability (Dcritical). The safe beach distance increments were determined using the PMP (probable maximum precipitation) and possible maximum wind speed. Based on this analysis, the optimal safe beach distance at the study site was identified as 202.2 m. Since the current beach distance (371 m) exceeds this value, the TSF was considered to be stable and satisfied the design safety factor of 2.0. However, this distance should be reassessed periodically to account for changes in the TSF geometry and other conditions. Furthermore, it applies only TSFs that are well-managed and feature a smooth, non-undulating beach surface with a consistent slope.

By Hengst B

The first production stage in any mining operation is the breaking of whole rock into fragments small enough for loading and subseq- uent handling. This primary fragmentation is most frequently accomplished with blasting. The maximum fragment size and the fragment size distribution have an influence on the subsequent loading, handling and crushing costs. It is assumed that total rock fragmen- tation in practice is a result of an interact- ing array of radial cracks from various blastholes. The representation for the basic pattern of radial cracks emitted from a particular blasthole is first described. A model is then developed simulating the interaction between blastholes at various burdens and spacings, and in simultaneous and delayed modes. The resultant picture is analysed and a fragment size distribution calculated. Methods of incorporating the existing faults in the rock are also discussed. Finally application of this model to various facets of mining are considered.

Jan 1, 1977

The classical view of island arc geology is that of a plutonic intrusive/volcanic extrusive complex, derived from the products of oceanic lithosphere subdu.ction and constructed on an over-riding plate of oceanic origin. Such an arc is typically backed by a marginal basin, formed by seafloor spreading of a more or less conventional kind. Various studies undertaken over the last decade have modified this picture and provided a new tectonic framework within which to explore for resources. Formation of Marginal Basins During the 1970s, seafloor spreading magnetic anomalies were identified in various marginal basins in the SW Pacific. In one of these, the South Fiji Basin, there was clear evidence of a ridge-ridge-ridge triple junction, although no kinematic relationship to the major plate boundaries was determined. Palaeomagnetic studies conducted in Vanuatu and Fiji, and in the Bismarck Archipelago of PNG provided evidence of large scale arc rotations relative to major plates. This indicated a general pattern of marginal basin formation by ridge-ridge-ridge triple junction growth together with island arc growth by trench-ridge-trench triple junction evolution. Formation of Primitive Island Arcs The triple junction growth model strongly suggests that at least a significant amount of arc evolution occurs by the initiation of subduction at a spreading ridge. This has predictive implications for the nature of island arc basement.

Jan 1, 1987

By M. Mortada

Nonsteady-state flow of slightly compressible liquids in porous media of non-uniform properties has been the subject of a number of recent studies. Most of these studies considered one-dimensional flow in systems with cylindrical geometry and with the properties of the porous medium varying with radial distance only. A. Houpeurt' studied the problem in its general form by examining the solutions of the differential equation, In Eq. 1, the permeability k and the porosity + are arbitrary functions of radial distance. Houpeurt concluded that, except for special forms of the functions k and the problem is not well suited for mathematical treatment, but lends itself to high-speed digital computation. Carslaw and Jaeger' discussed the solution of Eq. 1 for the case in which the permeability varies as a power of the radial distance. P. Albert3 studied analytically the pseudo steady-state solutions in systems of finite radial extent and non-uniform properties. He also obtained some numerical solutions for nonsteady-state flow using a high-speed computer. William Hurst4 examined the interference pressure drop due to a point sink located in an infinite system consisting of two permeability regions in series. In this case, the permeability changed at a radial distance r from the point sink. INTERFERENCE BETWEEN OIL FIELDS IN A NON-UNIFORM AQUIFER The interference pressure drop for oil fields located in an extensive aquifer of uniform properties is discussed in a number of places in the literature. This work considers the interference pressure drop for oil fields located in a non-uniform extensive aquifer comprising two regions of different properties as shown in Fig. 1. Region I of the aquifer extends from which is the radius of Oil Field A, to ar,,, where a>1. It has permeability k,, porosity +, and effective compressibility c,. Oil Field B is located in Region II which extends from ar, to infinity and has different properties from those of Region I, but the same formation thickness h. Region II has permeability k2, porosity and effective compressibility c2. The choice of this type of aquifer should not imply that it is of frequent occurrence. It is chosen primarily to illustrate the effects of non-uniform aquifer properties on oilfield interference. It is of interest to point out that a non-uniform radial aquifer in which both the permeability and the porosity are inversely proportional to the radial distance can be treated in a manner similar to a linear aquifer of uniform properties. The situation where only the permeability varies as the inverse of the radial distance is also amenable to analytical treatment. Both aquifers behave quite differently from a radial aquifer with uniform properties so far as the water influx and the interference pressure drop are concerned. METHOD OF SOLUTION The expression for the interference pressure drop in Oil Field B due to a constant rate of water influx in Oil Field A is developed in the Appendix. This is accomplished by the simultaneous solution of two partial differential equations describing the pressure in Regions I and 11. The two differential equations are solved subject to the following conditions: (1) at r,, the pressure gradient is constant; (2) at ar the pressure and

The basic flowsheet for the Kalgoorlie Nickel Smelter and Kwinana Nickel Refinery are discussed in relation to the mineral components of Kambalda produced nickel concentrates. Both processes handle violarite and pentlandite satisfactorily but the smelter has limitations where high copper recovery is required and in its ability to handle magnesia containing minerals. While the latter can be more economically handled at Kwinana the limitations of this tolerance are discussed together with problems arising from treating pyrrhotite, pyrite and precious metals by the Sherritt Gordon process.

Jan 1, 1973

By M. C. Irani

This survey was conducted to determine the magnitude of the methane emission and control problem in U.S. bituminous coal mines. The resultant data indicate that the emission rate for any given coalbed depends primarily on the coal production rate and on mine depth, as well as on the nature of the coalbed and the surrounding strata. An excellent correlation was found between (1) the methane emission rate and (2) the product of coal production rate and mine depth for the mines in the Pittsburgh, Pocahontas Nos. 3 and 4, and the Illinois Nos. 5 and 6 coalbeds. The highest total emission rates found in U.S. bituminous coal mines were as follows, in million cubic feet per day: Monongalia County, W. Va., 39; Marion County, W. Va., 30.4; McDowell County, W. Va., 13.1; Washington County, Pa., 11.9; Greene County, Pa., 11.4; Buchanan County, Va., 21.6; Jefferson County, Ala., 10.3; and Franklin County, Ill., 6.6.

Jan 1, 1972

A dual rate kinetic model consisting of differential equations is used to describe the kinetics of the adsorption of gold cyanide on activated carbon. The partial differential equation describing the material balance in the liquid phase is written for the special case of a fixed bed system. It is shown that an isotherm for the case of a closed system is required to yield accurate simulations of the equilibrium in a column of activated carbon. Whereas the equilibrium loading of gold cyanide is influenced by the free access of oxygen to the system, the traditional open batch stirred tank experiments, which are used to estimate parameters for the calculation of breakthrough curves for organic solutes, cannot be used here. The program solves this model at incremental heights in the column at a specific future point in time. Runge-Kutta and backwards- difference methods are used in the numerical solution. A Von Neumann method is used to show that the solution is stable. Kinetic parameters can be estimated either from batch experiments or from short column runs. In the latter method the column should not be long enough to contain the concentration wavefront for the system of interest. Results obtained from bench scale continuous column experiments were used to verify the computer model. Pure gold solutions as well as real leach liquor were used in both short column and countercurrent moving bed experiments.

Jan 1, 1987

By L. Mabiza

Keywords: Pyrometallurgy, furnace, sulphur, SO2, off-gas, Zimplats Zimplats embarked on an off-gas sampling campaign in December 2009 to characterise the concentration of SO2 in the off-gas from their 13 MW furnace. In excess of forty isokinetic samples (US EPA Method 8) were taken over seven days. The aim of the sampling campaign was to increase the understanding of factors that influence sulphur release from sulphide concentrates in an electric furnace. Specifically, detailed monitoring was conducted over various operating conditions to evaluate: ? The relationship between SO2 and feed rate (i.e. furnace power) ? The relationship between SO2 and ingress (i.e. furnace freeboard pressure) ? The impact of concentrate bed thickness on SO2 (i.e. blacktop) ? The effect of events such as tapping and converter slag return on SO2 ? The desulphurization reactions that take place in the blacktop A model was developed to predict the SO2 content in furnace off-gas for a particular set of operating conditions. In order to evaluate the impact of random, uncorrelated changes in concentrate composition, ingress and power (feed rate) on the SO2 content of the off-gas, a Monte Carlo approach was adopted whereby concentrate composition, power, and ingress were allowed to vary randomly around selected set-points. The model is primarily based on the release of ?labile? sulphur from the partial thermal dissociation of sulphide minerals with an over-stoichiometric sulphur content (e.g., Fe7S8 decomposition to FeS and S). The main findings of this work suggest that labile sulphur does not fully account for the measured SO2 content of the off-gas, and that limited oxidation of sulphides during melting also contributes to the SO2 content. Details of sulphide oxidation reactions in the furnace remain unclear, and it is currently not possible to provide a theoretical framework for the relationship between the extent of oxidation and variables such as temperature, gas composition, particle size, and residence time in the furnace black top. Through the use of the Monte Carlo model it was, however, possible to constrain the extent of sulphide oxidation to about 1%. This work has shown in practice what was expected in concept, that SO2 in off-gas is proportional to furnace power and ingress rate. The model that was developed consistently slightly under-predicts the quantity of SO2 generated. It also highlights that more work is required to better understand the analytical methods used to analyse for SO2 in off-gas.

Jan 1, 2011

By Shahzad Moosa, John M. Pitt

Statistical variability in rock strength can be evaluated by rapid, inexpensive ultrasonic pulse velocity measurements, an important consideration since rational probabilistic design in rock is often hampered by the cost of obtaining sufficient strength data. Experimental ultrasonic pulse velocity measurements on a set of 96 limestone specimens showed two distinct statistical populations. Both populations followed Weibull distributions, with flexural strength distributions paralleling pulse velocity distributions. Thus, experimental results suggest that strength variability may be predicted from a few rationally selected strength measurements and several ultrasonic tests. The correspondence between strength and velocity may then be used to develop an experimental transformation function, allowing the development of a statistical distribution function for strength based on the known statistics of velocity. Comparison of experimental groups with full sets of corresponding strength1 velocity data showed the technique offers a fairly accurate description of the statistics of strength.

Jan 1, 1986

By Neville G. W. Cook, John Kemeny

Utilizing the principles of elastic-brittle fracture mechanics, stress-displacement relationships are derived for an elastic solid containing a slip plane consisting of various configurations of collinear arrays of cracks. These models take into account the heterogeneous nature of slip planes as observed in the lab and in the field, and are relevant to the behavior of rock containing faults, joints, bedding planes, and fractures. All of our models exhibit slip weakening behavior, and show regimes of both stable (aseismic) and unstable (seismic) deformation. For given configurations of cracks and applied boundary conditions, our results show explicit ly the energy released through unstable crack growth. For instance, it is found that two equal length cracks growing together release more seismic energy than, say, a large crack growing into a small crack. Also, by allowing cracks to coalesce, our models show multiple stick-slip events without crack healing or other geochemical means.

Jan 1, 1986

Over the past 5 years, Bougainville Copper Limited has undertaken a number of major plant expansions to the copper concen- trate production plant at Bougainville in the North Solomons Province of Papua New Guinea. These projects have scheduled three additional ball mills, three new banks of Outokumpu flotation cells, four additional tertiary screens, two additional tertiary crushers, and more recently four high capacity 24.5m dia. tailing thickeners. The value of the five separate projects totalled more than A$100 million. A requirement of commissioning and bringing on line of each of these projects has been to minimise both operational losses and copper concentrate production. The commissioning has been successfully implemented by detailed planning of the commissioning work in conjunction with the plant operation. Planning has been detailed for each man involved on an hourly basis, highlighting the work required on each commissioning task that is to be undertaken and the time expected to achieve activity Completion.

Jan 1, 1986

By J. B. Wagner, B. Swaroop

The atomic ratios of oxygen to iron in zlarious corn-posilions of wustite in the vicinity oj the reported p to n transformation were determined in the temperalure range between 950" and 1250°C. For these conditions, equilibrium vacancy concenrations have also been determined at vatious mixtures of C0,-CO. The data show a continuos introduction of iron vacancies as oxygen is introduced into wuistite. No phase change is observed with in the wuustite field. However, the value of n in the isothermal oxygen pressure dependence, the weight change shifts to 1ower values as the concentration of oxygen increases. WUSTITE is a metal deficit semiconductor with the majority lattice disorder being cation vacancies. Using the notation similar to that of Kroger and ink,' the dissolution of oxygen into wiistite using C02-CO mixtures is represented by, WHere v+! denotes a cation vacancy with k effective negative charges and kh • denotes k electron holes, h -. Applying the law of mass action and assuming a dilute solution of defects in the wiistite yields Where the square brackets denote concentrations. Furthermore, when k = 2 the cation vacancies are completely ionized, and both the concentration of cation vacancies and the concentration of electron holes are proportional to the third root of the ratio, PC /PCo. This analysis is valid for one particular mael based on an ideal solution at elevated temperatures (in this case t > 1000°C). At lower temperatures, it has been shown2 that the pressure dependence of the electronic conductivity changes in a manner which suggests that vacancies interact with or "trap" electron holes. At the very high concentrations of defects, 10'' to 10" per cu cm, these simple models no longer hold. Recently Tannhauser,3 Geiger et a1.' Bransky and Tannhauser; and Hillegas' have found an apparent p to n transformation in wiistite in the temperature range of 950" to 1300°C. This junction was found by thermoelectric and electronic conductivity measurements. Geiger eta1.' found that the electronic conductivity became almost independent of the oxygen pressure at high temperatures and oxygen concentrations within the experimental error. In addition, Vallet, Raccah, and coworkers6'7 have reported the existence of several "varieties" of wiistite which were reported to exist within what was heretofor presumed to be a

Jan 1, 1968

By John B. Hastings

Chief among the varied problems facing the mine-manager is that of vein-structure and origin, which is highly important as a guide to successful discovery and development. If metalliferous deposits can be traced to the intrusion of waters along definite lines, then is there something tangible for him to study. But when he is told by a geologist that metalliferous deposits are due to some other cause, like magmatic differentiation, while not perhaps discouraged, he is impressed with a vague sense of new worlds to be conquered. In 1897, while manager of the War Eagle mine at Rossland, B. C., where, previously, the ore-bodies had not been supposed to occur as veins, I suggested to visiting members of the Canadian Geological Survey that, perhaps, the structure of the pyrrhotite-bodies of that mine, which had been proved to occur along well-defined fractures, might throw some light on the similar deposits of Sudbury, then held to be basic aggregations from the original magma. Later investigation has at least provoked discussion of this last deduction, questioning its correctness and suggesting that, perhaps, the deposits are actual veins. The following extract from a paper by J. E. Spurr' is quoted, because it sets forth Mr. Spurr's theory of the quartz-occurrences of the Silver Peak mines, which will be discussed in this article. It also graphically describes the attendant geological features. "The Drinkwater group of mines, which is the most important part of the Blair gold properties, and which has produced practically all of the million dollars' worth of ore, as above stated, may be taken as typical of the gold-veins which, though widespread and numerons, show a wonderful similarity of character. On the surface two adjacent veins outcrop, the Crowning Glory and the

Jan 1, 1906

By Jheison Lopes dos Santos

Sistemas de blindagens multicamada têm sido estudadas pela sua ótima capacidade de proteção, devido à combinação de diferentes materiais que constituem seus componentes. A primeira camada é constituída por um componente cerâmico, cujo objetivo principal é fragmentar o projétil, dissipando grande parte de sua energia. A camada intermediária compreende tecido de aramida ou matriz polimérica com fibras naturais, que também dissipa energia tanto do projétil quanto do cerâmico e opera retendo os fragmentos cerâmicos. Neste trabalho, pela primeira vez, o sistema cerâmico Al2O3-Nb2O5-LiF tem seu comportamento dinâmico avaliado balisticamente. Este composto cerâmico foi submetido a impactos balísticos de projéteis 7,62 mm, para testar este cerâmico do primeiro componente de um sistema de blindagens multicamadas. Os resultados obtidos mostraram que a adição de LiF ao composto cerâmico Al2O3-4%p. Nb2O5, sinterizado a 1.300 ºC, obteve desempenho comparável ao obtido por Al2O3-4%p. Nb2O5 sem LiF, sinterizado a 1.400 ºC. Os resultados revelaram que, ao serem submetidos a esses impactos balísticos, apresentaram valores comparáveis aos esperados de pressão e de velocidade de partícula na interface de impacto projétil-cerâmico. Dessa forma, a adição de fluoreto de lítio ao composto cerâmico alumina-nióbia possibilitou a redução em 100 ºC na temperatura de sinterização, sem comprometer o desempenho balístico e exibindo parâmetros de choque comparáveis, e redução de custo.

Oct 30, 2017

By H. R. Banks

PONDERING the action of a ball mill employed to carry on comminution from the product of fine crushing to the point where flotation, cyanidation or other concentrating methods are applicable, one wonders where this machine had its origin. When Henry Wadsworth Longfellow reiterated the observation of that older poet Von Logau, and wrote “Tho’ the mills of God grind slowly, yet they grind exceeding small," he probably visualized the type of mill that was employed in medieval times for the production of flour, and can therefore not be considered as having gained admittance to the fraternity of millmen, on behalf of, either poet. He does, however, make it quite clear that the efficiency of even these Mills of Destiny was low, and to that extent voices a characteristic of fine-grinding equipment that has been the bugbear of all metallurgical engineers. PRINCIPLE OF BALL MILL The principle of the ball mill probably is traceable to the action along the coast line of oceans or lakes, or in the mountain streams. Wave action on rocks and pebbles, close to the shore line, gives the potential energy to the grinding media, which, in their rolling and tossing, break down the sands, and the resultant "pulp" carried out from the shore is classified and settles out in sizes ranging downward to extremely fine mesh. Surely, these are the real mills of God. If we allow our minds to dwell on the total tonnage that has been handled in this way, adding to the sands and clays at present in existence those enormous tonnages of metamorphosed sedimentaries, we sense the vast quantity of rock that has been so reduced. Is it not likely, therefore, that the answer to the beginnings of pebble milling, and eventually ball milling, came through the perception of this natural mill by one of our engineering forebears? METHODS OF GRINDING The relatively low return of mechanical input to a ball mill, in terms of useful work accomplished, has always intrigued me immensely. The potential energy absorbed by a ball raised to a high point in its path is apparently released with a very small return in useful work. Is the ball mill the most desirable means of fine grinding, or is there some other method through which we can recover a reasonable amount of the energy input in terms of useful work? Some years ago, on his retirement, H. W. Hardinge was questioned as to his thoughts on the trend of fine-grinding equipment. As I recall, his prophecy at that time was that the most likely field for improvement in grinding media would be toward a charge of flat disks with some mechanical scheme to cause them to slide over each other. Undoubtedly, Mr. Hardinge went through the usual experience of the muller and bucking board, the customary introduction of the budding millman to his calling. I have often wondered, but have never got beyond the wondering stage, as to the

Jan 1, 1945

By Y. C. Kim

Three rock fracture parameters define three points along empirical cumulative size (of rock fragments) distribution curves at Inspiration copper mine. These parameters are: rock quality designation (RQD), longest piece of intact rock and % of +1 inch. The RQD and the longest piece are actually measured on drill hole cores whereas % of +1 inch is visually estimated on the same cores. These size distribution curves are then utilized in a variety of engineering applications such as; slope instability analysis, blast design, copper recovery prediction from heap leaching, etc. The three rock fracture parameters for two dominant rock types (porphyry and schist) in the pit had to be estimated separately due to different rock fracture characteristics.' Both ordinary kriging (OK) with measured hard data only and soft kriging (SFTK) with hard and soft data were used to estimate each parameter. The cross validation of 100 measured sample values served as the-basis for evaluating the potential benefits of SFTK over OK. Use of subjective information during SFTK required obtaining (prior) probability distributions for all three rock fracture parameters. These distributions were obtained from the typical rock fracture parameters observed in drill hole cores falling inside the geologically interpreted structural domains with varying degrees. of fracture intensity and also through direct mapping along the exposed bench faces. Two different types of soft information were developed for this study; namely, Type "A" and Type "K". The former is based on sample distribution of fracture parameters whereas the latter is based on two step estimation designed to narrow the range of variation observed in data Type "A". The results of the cross validation were analyzed using a linear regression of actual values versus estimated values. The results confirm an improvement on the estimates made by SFTK with hard and. both types of soft data over those produced by OK with hard data only.

Jan 1, 1989

By H. S. Eisner, J. -P. Leger

Published South African fatality and injury statistics for mine accidents are investigated. Where possible, the South African statistics are compared with those of other countries. Commonly held attitudes to the South African accident experience are also examined. Analysis reveals that the published accident rates present a more favourable picture than the rates actually experienced by underground workers. In addition, fatality rates in South African underground gold and coal mines do not differ significantly and have remained stable over the past decade, although the injury rates have decreased rapidly. No evidence was found that this is due to improved safety, and a comparison with international fatality and injury data suggests that it is not. The underground fatality rate for South African coal mining is, on average, eight times that of the UK, four times that of all EEC coal mines taken together, and over double that of the USA. Despite the unreliability of the injury rates, much has been made in South Africa of their rapid decline, an emphasis that appears to be misplaced.

Jan 1, 1988

By R. J. Dippenaar, W. H. Van Niekerk, D. A. Kotze

It is generally accepted that alkalis attack and degrade the carbon fraction of coke in a blast furnace. Because South African cokes contain relatively high amounts of ash, the influence of alkalis on the ash fraction was investigated, the carbon and ash fractions being separated and studied separately. Potassium engenders the formation of liquid, with a resultant expansion of volume and a decrease in the strength of the coke ash. However, the ash did not play a significant part in the degradation of the coke, and it was found that the catalytic effect of potassium on the Boudouard reaction was the overriding factor in the coke-degradation process.

Jan 1, 1986

By Michael Nicol

A quantitative description of hydrometallurgical operations depends primarily on an understanding of the kinetics of the chemical and physical processes that take place in such operations. This paper will review the current status of kinetics in hydrometallurgy in terms primarily of the author’s experience in diverse unit operations. A distinction will be made between the popular use of models based on limited kinetic data and mechanisms which reflect the actual elementary chemical or physical steps involved in an overall process. Several examples will be used to illustrate this distinction. The proliferation of published papers of declining quality has resulted in the bequest to later researchers of outdated and often incorrect conclusions by a process that is often difficult to reverse. Some examples of these mysterious concepts will be examined and hopefully reversed. Finally, several examples will be given that identify significant gaps in our understanding of the kinetics of important processes that remain unresolved.

Jan 1, 2014

By William H. Grady

FactoRs entering into the market value of coal are its grade, and the cost of labor, material, and capital. Reduction in these costs cannot be expected in the future, and it therefore follows that greater economy in their use must be accomplished if it is desired to hold the present markets; more tons of product per unit must be obtained in a given time from both men and material, and means devised for increasing the percentage of lump and domestic sizes. This paper is presented for the purpose of inviting discussion and stimulating thought in regard to the influence of methods of procedure and plans of mining, upon the quality of coal, the economical use of labor, material, and capital, and the cost of production. Even the better methods of the present day leave much to be desired, and a review of the reasons and necessities for their use may be of value in pointing the way to more economical mining. The writer does not attempt to discuss untried methods of mining, but rather to state clearly and concisely a summary of the results obtained under the several methods of procedure that have been adopted. In some instances a high degree of concentration has been effected, and comparisons of the results obtained with the results of lesser degrees of concentration form an interesting study. It is believed that much may be accomplished in improving the quality, and in the more economical use of labor, material, and capital, by concentration, and it is to concentration in particular that your study is invited. The writer, who has occasion to visit many mines, most of which are under different management, has been inclined to the opinion that improvement has been retarded in many ways and that mining has not as yet been entirely freed from the early-day practices which were forced upon it. Practices resulting in lack of concentration, to which reference will be made, are: The absence of robbing, necessitating the frequent interposition of barrier pillars; lack of system and of proper supervision, resulting in losses of life and coal. These practices could not well be

Jan 1, 1915