Search Documents

Treatment of Gold-Bearing Oxide High-Clayish Ores The analysis of modern state of treatment technologies of gold-bearing oxide high-clayish ores showed that the main method applied both in Russia and abroad is sorption leaching and heap leaching with preliminary granulating. The use of any of these processes depends on many factors including average gold grade, site and area of a deposit, clay content and etc. The two processes are often used at the same deposit. In this case, rich ores are treated according to a plant technology, poor ores and ores with cut-off grade are treated with heap leaching. Both technologies possess their own advantages and disadvantages. A plant technology is characterised by high recovery with higher capital and operating costs at the same time. Lower recovery during heap leaching can be attributed to an unstable quality of ore granulating due to both technological and weather conditions. In case of poor high-clayish ores, the use of a combined ore treatment is possible and profitable. It includes the ore disintegration and a clayish fraction treatment by using a heap leaching. In this case, capital and operating costs reduce in comparison with a plant technology and the recovery increases in comparison with heap leaching. Sometimes it reaches a plant recovery. These ores possess special features. As a rule, gold is associated with ore fraction, so it liberates readily from the association with ore ore-forming minerals even with relatively coarse grinding explain their high solution velocity in a hydrometallurgical cycle. Clayish ores are characterised by high sorption activity to gold cyanide complexes as well as extremely low velocity of filtration and thickening of slimic pulps at the same time. The paper describes the options of oxidised high-clayish ores treatment, such as, a plant technology (PIP and CIP process), heap leaching and a combined technology (disintegration, heap leaching of a crystalline fraction and sorption cyanidation). The obtained results were used when selecting a treatment technology and during construction of a plant for oxidised high-clayish ores treatment.

Sep 13, 2010

By P M. Ashley, M Muller

Massive sulphide ZnPbCuAg deposits in the Halls Peak district of north-eastern NSW are of the submarine volcanogenic type. They occur as a number of small lenses enclosed in shale and epiclastic rocks, in association with felsic volcanics of early Permian age. The deposits crop out in steep terrain near the eastern edge of the New England plateau. Local relief of 700 m and slopes up to 50¦, coupled with an average annual rainfall of about 1000 mm has led to local severe land sliding in previously mined areas. Mining from the 1890s to the 1960s has also resulted in unconstrained drainage of heavy metal-bearing acid waters and physical dispersion of finer grained metalliferous sediments into the Chandler River. Waters emanating from the main landslide area, from underground workings and draining scattered unconfined stockpiles of massive sulphide ore have pH values of 3 to 4 and contain elevated contents of Zn, Cu, Pb, Cd and Fe. Chemical mobility of heavy metals into streams occurs under low pH conditions with Zn exhibiting the greatest dispersion. Increasing pH conditions downstream cause precipitation of Fe phases and coprecipitation of heavy metals. Physical dispersion of secondary metal-bearing minerals into the surrounding creeks results in wt% levels of Cu, Pb and Zn in the stream sediments thereby exceeding the regional background values by one to three orders of magnitude. Areas disturbed by mining are characterised by lack of vegetation, biomagnification of heavy metals into algae and lower plants, and partial recolonisation by metal-tolerant grass species (Snow Grass, Poa sieberana; Blady Grass, Lomandra longifolia). Rehabilitation measures will have to include landslide stabilisation, impervious sealing or removal of sulphidic materials, import of topsoil and revegetation by metal-tolerant, slow-growing native grass and shrub species.

Jan 1, 1995

By M K. de Brodtkorb

Late Cenozoic ore deposits æPrecious and base metal-richÆ and æSn-Ag-polymetallicÆ ofNorthern Puna are associated either with Middle Miocene dacitic volcanic-dome complexes or with Upper Miocene dacitic-andesitic volcanic centres.   Ore mineralogies and their related volcanism are discussed. A strong link is observed between the Cenozoic Metallogenetic Province of Northern Puna and the Southern Bolivia Tin and polymetallic provinces.  

Jan 1, 1995

By A Scott, A C. Torrance

Explosive properties are typically summarised on manufacturers’ technical data sheets. The values most commonly referred to are the density of the explosive, the energy it is expected to release and its velocity of detonation (VOD). This information is provided so that blasting engineers can make a considered selection of the most appropriate product for a particular blasting application. As an example, a lower VOD and energy product is better suited to blasting in soft, weathered material. In contrast, blasting hard, competent material generally requires a higher energy product which is usually characterised by higher VOD and energy.A review of the technical information supplied by manufacturers shows a wide range of methods of reporting both VOD and energy. A meaningful comparison between products from different suppliers is almost impossible to achieve based on these data. For a similar product, one supplier may have a VOD range of 2800–5800 m/s and a relative bulk strength of 1.34 while a corresponding product from another supplier is quoted with a VOD range of 3100–6300 m/s and a relative bulk strength of 192. In the end, the user has to rely on the level of confidence available from their experience using a certain type of explosive in the type of rock to be blasted at that mine.This paper reviews some of the technical and supporting information that is freely available in the literature and attempts to provide some clarity as to what the numbers that are reported actually mean for the end user. An alternative approach to the evaluation of the suitability of a given explosive to blast a given rock mass is proposed.CITATION:Torrance, A C and Scott, A, 2015. What is relative about energy?, in Proceedings 11th International Symposium on Rock Fragmentation by Blasting, pp 447–454 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 24, 2015

By Y Zhang, X Yi, J Deng, J Zhao, J Song

To obtain the combustion process of a coal fire, a high-temperature programmed system is used to simulate the spontaneous combustion process of fresh and weathered coals. Both fresh coal and weathered coals from Liulingou, Inner Mongolia, China, were used to examine the gas releases from constant temperature to 650°C. Moreover, gas release rate was used to calculate the characteristic temperatures. The results showed the coal samples lasted to oxidise under high temperatures, and released plenty of heat to assist self-heating. The humic acid contained inside the weathered coal went through thermal decomposition as temperature rising and generated gases. This resulted in increasing concentrations of CO, CO2, CH4, C2H4, and C2H6. Changes of CH4, C2H4, and C2H6 gases were also similar. For weathered coal, concentrations of three gases were kept in balance under low temperatures, increased rapidly with the temperature rising. After the ignition temperature, the concentration of hydrocarbon was reduced. In addition, particle size affected coal spontaneous combustion. Coal samples with less than 0.9 mm in size produced much more CO than others. This study has examined coal spontaneous combustion caused by coal weathering during mining and its impacts on self-heating and pyrolysis in the process.CITATION:Deng, J, Song, J, Zhao, J, Zhang, Y, Yi, X and Zhang, Y, 2017. Gaseous products of weathered coal during the process of spontaneous combustion, in Proceedings Australian Mine Vent Conference 2017, pp 79–84 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 28, 2017

By H Peiming, W Sakurai, Y Matsuhisa, H Shimazaki, Y Fukahori

Oxygen and carbon isotope mapping of hydrothermally altered wall rocks was applied to the blind ore prospecting of skarn Pb-Zn deposits in the Kamioka mining district, centralJapan, where the wall rocks consist of heterogeneous rock units. The results (d18OSMOW of -1.1 to +17.3ë and d13CPDB of -5.0 to +4.8ë for limestones, and d18OSMOW of -0.8 to +12.5ë for silicate rocks) show isotope zonations of the wall rocks with depletions in heavy isotopes (18O and 13C) toward the centre of mineralization. The isotope zonations likely formed by interaction of migrating thermal waters with the wall rocks during the mineralisation. The heavy-isotope depleted zone, which indicates higher temperature or higher water-to-rock ratio of the system, occurs in the footwall of a fault, indicating that the fault played the role of main conduit of the ore-forming fluids. A heavy-isotope depleted zone was also found in an area, where significant mineralisation had yet to be identified. Recent drilling in this area has intercepted a zone with a grade of 16.0% Zn, 0.8% Pb, and 35g Ag per metric tonne at a depth of about 360 m.

Jan 1, 1995

By A C. Carter

The increasing need for alternative energy sources as well as the less clearly perceived but equally increasing demand for non-fuel minerals has resulted in extensive exploitation of near surface deposits. Many such deposits have depleted reserves and mining corporations are increasingly turning their attention to deposits, with thicker overburden, less competent materials, steeper dips and more complex structure. These deposits pose mine layout, geotechnical and equipment problems as well as the need for large outputs to achieve economies of scale. The paper describes the major problems and operational solutions for such deposits using continuous excavation and transport equipment.

Jan 1, 1980

By J J. Robinson, B N. Noller, D M. Franks, G McIlveen, D R. Mulligan, A T. Soro

Fiji is a developing nation that relies on economic opportunities and recently experienced increased interest in mining and mineral exploration that could potentially benefit the country. In order to understand the lack of, or failure of, past rehabilitation activity, this study applied risk-based approaches to identify the significance of activities or actions applicable to rehabilitation of post-mined land in Fiji. This involved the use of qualitative, semi-quantitative and quantitative risk methodologies, which allow assessment and evaluation of mechanisms of mining administration and practices, as well as rehabilitation methods and approaches that have been applied successfully to post-mined land in other countries with similar environmental and climatic conditions and social and regulatory environments to Fiji. The semi-quantitative risk-based technique followed an earlier approach applied for mine closure by Laurence (2006) and provided a systematic way to assess and evaluate rehabilitation issues that were identified on-site by assessors, who were able to estimate through expert judgement the likelihood and consequences based on conditions, situations observed and circumstances on-site. It demonstrated that experienced professionals could reproducibly score likelihood and consequence of selected items relevant to mine site rehabilitation, despite the wide-ranging issues identified during post mined land rehabilitation. The quantitative risk-based approach was adopted as a method to determine environmental bonds for mining projects using the assessment of consequences and likelihood results from the semi-quantitative approach, then incorporating unit cost of materials or activities for local conditions required to fully rehabilitate each issue at net present value. Application to both an abandoned mine site and a newly opened mine site in Fiji showed that this approach could provide a risk-based estimate for rehabilitation cost and demonstrated potential applicable in the Fijian context.CITATION:Soro, A T, Mulligan, D R, McIlveen, G, Franks, D M, Robinson, J J and Noller, B N, ?2014. Risk-based Rehabilitation Strategies for Post-mined Land in Fiji, in Proceedings Life-of-Mine 2014 , pp 165–180 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 16, 2014

By Secis R

Northparkes Mines is located 27 km northwest of Parkes in central west New South Wales. The E26 deposit is the largest of all defined to-date in the area, containing over 50 per cent of Northparkes' resources and reserves. The combination of orebody geometry and geotechnical characteristics provide ideal conditions for the block cave method of mining of the first lift of the E26 deposit. Although this method of underground mining has been used for decades in other countries, E26 is the first example of its use in Australia. Copper-gold mineralisation at Northparkes is related to quartz monzonite porphyries that have intruded coeval Ordovician volcanics. Sulphide mineralisation (dominantly bornite and chalcopyrite) occurs principally within quartz veins associated with the porphyry intrusions. Veining and mineralisation occurs within both the porphyries and volcanics. To optimise the recovery of a resource within a block cave mine, a good understanding of the rock mass and ore flow characteristics is essential. These have a significant effect on the fragmentation size and range and on parameters such as the drawzone spacing, height of interaction zone and dilution entry. As ore within a cave column moves and mixes when material is extracted from below and beside it, the total reserve extracted is unique for a particular draw strategy/production history. Until recently, reserve estimates and production schedules were based on the application of a uniform dilution factor to the tabulated resource. However, given the complicated flow and mixing within the cave column, such a simplistic approach can produce erroneous results. Northparkes Mines purchased and sponsored the upgrading of a commercially available software package specifically developed for modelling removal of resource blocks under caving conditions. This Gemcom software package, PC-BC, has been used for mine planning, reserve reporting and production scheduling. Blocks within it resource block model are converted into a mineable reserve taking into account mixing and dilution during depletion of drawcolumns (by horizontal and vertical redistribution of insitu tonnage and metal). As PC-BC gives individual drawpoint reserves, it can be used during the planning stage to determine whether or not a drawpoint is economic. For E26 Lift I an optimal layout containing 130 drawpoints has been developed based on a nominal 14 m x 14 m drawpoint spacing and a 0.8 per cent equivalent copper (eCu) cut-off grade. Annual production schedules for the depletion of material above this layout are based on: I higher tonnage removal from higher grade drawpoints in the early years of production; and 2 drawpoint production rates proportional to the total tonnage of orr remaining in the drawcolumn up to the optimal height of draw tier subsequent years. The reserve is tabulated from the combined annual production schedules over the life of the lift. The pre-mining reserve for Lift I of E26 is 26.6 Mt at 1.43 per cent copper and 0.39 g/t gold, of which approximately 23.4 Mt will be extracted under caving conditions. By scaling down the annual production schedule, weekly drawpoint tonnage calls can be given for short-tern production scheduling.

Jan 1, 1997

By A Giblett, T J. Napier-Munn

This paper presents the results of a test program evaluating the impact of sample mass on the precision of Stage 1 gravity recoverable gold (GRG) test results. Four ore types ranging from low to high GRG contents were the focus of testing. A method to predict the influence of sample mass on the confidence limits of the Stage 1 GRG test result is demonstrated. The results suggest that although small samples tend to overestimate Stage 1 GRG in samples with low gravity gold values due to higher mass pull to concentrate, such samples can give acceptable estimates of the Stage 1 GRG content with focussed sample treatment protocols.CITATION:Giblett, A and Napier-Munn, T J, 2015. Measuring the influence of sample size on the precision and accuracy of gravity gold estimation, in Proceedings MetPlant 2015, pp 367–372 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 7, 2015

By J Lööw, M Nygren, E Segerstedt, L Abrahamsson

The purpose of the paper is to discuss socially sustainable development in the mining industry and the communities surrounding the mines. The discussions are based on results from a prestudy and literature review on mining and sustainable development conducted during 2013–2014 at Luleå University of Technology in Sweden. ‘A social licence to mine’ is important for the mining industry, but the social dimension is a relatively underdeveloped dimension when it comes to sustainable development in general and the mining industry in particular, one reason probably being the lack of effective methods for capturing social impacts. The mining industry and the surrounding communities face many challenges that provide both possibilities and obstacles to socially sustainable development; eg aspects such as gender, work conditions and cultural aspects. For example, a strong mining workplace culture and community identity can create strong cohesion but also lead to excluding certain groups, rejecting new ideas and reinforcing obsolete values. Other challenges include recruitment, as well as health and safety in relation to an increased use of fly-in-fly-out, contractors and automation of mining. Some challenges relate to the effects of fluctuations in the mining market. There is a lack of research that links attitudes, policies and activities within companies to their impact on the wider community, and vice versa. Future research should also include the development of methods and indicators for social sustainability relevant for mining – in other words: how do mining companies get ‘a social licence to mine’?CITATION:Abrahamsson, L, Lööw, J, Nygren, M and Segerstedt, E, 2015. How to get a social licence to mine, in Proceedings Third International Future Mining Conference, pp 125–136 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Nov 4, 2015

By Z Li, A Fleming, S Webster, L Snyman

Northparkes Mines is developing the E26 Lift 1 North block cave mine, which is adjoined to the existing E26 cave. This paper presents the challenges and methods used to design the ground support system during the feasibility study. Calibrated numerical simulations were conducted to forecast the levels of rock mass damage and to assist the ground support design. Generally, low to moderate rock mass damage is forecast on the extraction and undercut levels. However, poor ground conditions are expected in sheared and preconditioned zones adjacent the existing E26 cave. This will result in difficult ground conditions during development, undercutting and drawbell establishment. Empirical assessments were conducted to make sure the ground support system is able to handle both the static and dynamic load demand, and to accommodate the deformation values. A safety factor of 1.5 was used for the designs to cover the unknown rock mass conditions and potential for larger seismic events. Learnings from the E26 Lift 1 and SLC, also mining against the E26 cave, are incorporated into the design. Fibrecrete, mesh and resin bolts / mechanical anchors are the primary ground support system used, and cable bolts for intersections with a span of more than 6 m. Following undercutting and during the operating life of the mine, secondary ground support and rehabilitation will be required.

Mar 15, 2021

A variety of environmental financial assurance instruments are in use across the globe. This paper reviews some of the more common forms currently used in New Zealand, Canada, USA and Australia. In order to be effective, the form of assurance instrument selected needs to be aligned with a countryÆs specific environmental objectives, geography and political climates and also allow for a healthy investment climate. Industry and Government both have a strong interest in ensuring financial assurance criteria meet these objectives.   Recently Golder Associates have been involved in financial assurance assessments for various mining and quarrying companies across the South Island. Multiple stakeholders and regulatory agencies can make the setting of effective and fair financial assurance criteria a complex exercise for both regulatory agencies and industry.  The author relates experience gained in assessing financial assurance under current New Zealand legislation (RMA, 1991), effectively dealing with issues such as ôdouble bondingö and overviews potential approaches to dealing with transitional issues for historic and existing operating mine sites.

Jan 1, 2006

At its Port Kembla Plant, Australian Fertilizers Ltd. manufactures Superphosphate and other Phosphatic Fertilizers.Phosphate Rock is received from Nauru and Ocean Island. It contains water insoluble tri-calcium phosphate, Ca3(PO4), to an average of about 86%.It is crushed to pass a 20-mesh x 26-gauge woven wire screen, and is then treated with Sulphuric Acid to give the product Superphosphate with a water solubie prosphate content of 20.5%, estimated as P205. Sulphuric Acid is made by the chamber process. Three sets of chambers are installed, on,e five-chamber set of timber frame work construction and the other two of more recent date, each with four chambers of steel frame work construction. A fourth chamber set is now being built.Sulphuric Acid is delivered for use. at about 70% H2S04, the strength depending upon the rock used.Mixing - The crushed rock and sulphuric acid are mixed in weighed and measured proportions respectiyely in one-ton batches in revolving cast fron pans, mechanically stirred. The mixture, while still slurry, is run into a closed horizontal concrete cylinder (den) where, with the completion of the chemical reaction, it rapidly sets as Superphosphate.Approximately fifty batches are required to fill a den. When full, the den is opened ana the mass cut out by a revolving cutter. The product is elevated and conveyed to any required position in the Superphosphate storage shed, where it dries and matures.Despatch - As required, the Superphosphate is excavated by a two ton grab and conveyed from the matured "dump" to bagging units where it is disintegrated, screened and bagged at twelve bags to the ton. The filled !Jags are closed by machine and delivered to railway trucks immediately.Mixed Fertilizers - Sulphate of Ammonia, Sulphate and Muriate, of Potash, and Blood and Bone are mixed with Superphosphate and thoroughly incorporated to give different mixtures to meet special plant requirements.

Jan 1, 1938

In this, the centennial year of The Institute, it is appropriate that the history of the Institute in New Zealand be recorded. I am indebted to John Dew, compiler of the centenary publication æMining People A Century - Highlights of the First Hundred Years of The AusIMM 1893 - 1993Æ, for many of the early historical references in this paper. In July 1892 the Amalgamated Mining Managers' Association of Australasia held its annual conference in Ballarat, Victoria, Australia, when it was suggested by the delegate from Broken Hill that the time had arrived for the establishment of an Australasian Institute of Mining Engineers. Resulting from this meeting at Ballarat, a Provisional Committee was formed at Broken Hill with the object of forming The Australasian Institute of Mining Engineers having the following objectives. To promote the arts and sciences connected with the economic production of the useful minerals and metals, the welfare of those employed in those industries by means of meetings for social intercourse, and the reading and discussion of professional papers with the circulation by means of publications amongst its members and associates the information thus obtained.

Jan 1, 1993

This paper examines the integration of risk management into the design philosophy of a project to produce a æzero baseÆ design, which targets a minimum capital cost without adversely impacting upon the operating costs of the resulting plant. The concept of zero base design is similar to zero base budgeting and the desired outcome is to improve the economy of the design. Design requirements are questioned in an attempt to distinguish preference, prejudice and myth from necessity. Where doubt remains a risk management approach is adopted. The key to risk management is not risk elimination, but risk mitigation. Risks can be managed by designing them out, but often this involves considerable capital cost penalties and it is more attractive to accept the risk and design an operating strategy to deal with it. The paper discusses this with some examples of real world risk mitigation strategies that resulted in substantial capital cost savings. The paper concludes that as much as a quarter of project capital costs can be eliminated by a combination of zero base design and carefully crafted risk mitigation strategies.

Jan 1, 2004

By I Halikia, T Lilis

Lead smelter slags, produced during the last century in quantities ofmillions of tonnes at Lavrion, Greece, have recently been used in theconstruction of Lavrion harbour. In order to evaluate the geochemicalstability of slags and assess their environmental impact on the aquatic environment, a detailed study on lead leachability under conditions that simulate subsea environments was carried out. Leaching experiments using magnesium chloride solutions under mild temperature conditions were conducted and the effect of main parameters on lead solubility was examined. Furthermore, kinetic analysis on the experimental results was performed. Based on the results, lead dissolution increases by increasing the reaction time, magnesium chloride concentration in the bulk solution and temperature, whereas it decreases by increasing the slags particle size. The reaction order was found to be approximately 0.5 at the initial extraction stages and the activation energy value from the Arrhenius plot was estimated to 29.8 KJ/mole. Diagnostic equations applied to all the experimental data reveal that lead dissolution is mainly controlled by diffusion through the ferrosilicate matrix of the slag particles, which is in accordance with the low activation energy value. The JanderÆs diffusion equation [1-(1-a)1/3]2=kt was proposed as the equation representing the reaction of lead dissolution from slags using magnesium chloride solutions.

Jan 1, 1999

The quantity, quality and detail of project information increases as project development moves from initial scoping level studies, to pre-feasibility and option studies, then definitive feasibility study and basic engineering stages. Data accumulates from metallurgical and vendor testing from various ore types and composites carried out to assess physical characteristics and processing responses of the ore. Over-simplification or assumptions about the orebody for flow sheet development can result in using æaverageÆ information for design. Use of averages of data can have adverse implications on sizing of plant and equipment and consequently on definitive study cost estimates and detailed engineering. In practice, the treatment plant workforce is faced with managing a non-homogeneous or non-average orebody in a variable operating environment that is not always predictable. This paper presents views on how to use the data generated from testwork in project development and for process and plant design. Examples of process design risk or operating risk are presented. Although this topic is applicable to the broad range of process design decision making in the minerals industry, this discussion references comminution, flotation and extraction processes for gold and base metal ores.

Jan 1, 2004

This paper presents a resume of the likely profitable mining areas in New Zealand. It covers the main 'minerals' û gold, coal, oil and gas, ironsands, silica sands, clays, lump silica, perlite, sulphur and limestone. The potential mining projects are depicted on a map of New Zealand together with the associated works involved in hydro-electric and geothermal power development requirement for industries such as aluminium, nickel ferro-silicon or silicon carbide production. The political need for mining development to relieve unemployment and reverse emigration is stressed. A case is made for exploration and development assistance by government in labour re-training, energy supply and transportation, if the tremendous losses through lack of use of human and industrial resources are to be rapidly reduced. These losses are currently estimated at $10 000 000 000 per year. The need for protection of the sanctity of contract and security of tenure of all mining operations is touched on briefly as is the need for a reasonably balanced view on conservation. However, as both of these are more properly covered in papers on the legal and environmental aspects of the mining climate, they are presented here only as they affect the economy of the mining operation. Finally, there is a brief comment on the mining climate in New Zealand relative to Australia and South Pacific mining operations.

Jan 1, 1990

By C Ferron, R Massimi

There is approximately 240 000 tonnes of arsenic trioxide dust stored in underground chambers at the Giant Mine in Yellowknife, Northwest Territories, Canada. To minimise the amount of arsenic released to the environment, a number of closure alternatives are being considered. One alternative is to mine the dust, encapsulate it and dispose the mixture into a secured landfill. The potential leaching behaviour of the encapsulated dust was examined in a laboratory programme. Cylinders of arsenic trioxide dust mixed with cement/aggregate and bitumen were prepared. The dust content varied between zero per cent and 60 per cent of dry weight charge. The cement cylinders were tested for strength, leaching and freeze/thaw durability. The bitumen cylinders were tested for leaching only. The results show that increasing dust content in the cement cylinders negatively impacts the strength and the freeze/thaw durability of the samples. A contributing factor is likely the high water to cement ratio in the samples. Leaching tests show that arsenic release from cement cylinders increase with increasing dust content. Arsenic concentrations in leachates do reach a plateau at a certain point in time. Arsenic release from bitumen cylinders was significantly lower compared to cement cylinders. Arsenic concentrations eventually were less than 0.05 mg/L in samples containing up to 25 per cent arsenic trioxide dust. Arsenic leaching rates were found to decrease by an order of magnitude over the course of three months.

Jan 1, 2003