Search Documents

"Here, we statistically proved that giant earthquakes occur at lunar phases specific to each subduction zone. Enough attention during the lunar period, especially when seismicity is occurring, will significantly reduce damage from giant earthquakes. Two case studies in which giant earthquakes occurred after seismicity in dangerous lunar phases were discussed, and the mechanisms underlying why giant earthquakes occur around neap tide at some subduction zones were explained by prohibition of giant earthquake occurrences due to high strain rates at spring tides at N–S subduction zones. The prohibition was statistically proven for uniaxial creep tests on Inada granite with slight stress disturbances. INTRODUCTIONIn previous studies, we showed that from 1900 to present, giant earthquakes (MW ??8) have occurred at lunar phases specific to each subduction zone (Fujii & Ozaki, 2012; Ozaki & Fujii, 2012). Although we later proposed a possible reason for these occurrences, the study lacked statistical analyses (Fujii et al., 2013). Therefore, this paper shows the results of such statistical testing and presents case studies for the Tohoku 2011 and Chile 2014 events, thereby providing updated data on giant earthquakes through July 2014."

Jan 1, 2015

By A. V. Lyamin, J. P. Wang, A. J. Li, Z. Qian

"For general stability analysis of rock slopes, rock mass strength and rock mass disturbance are definitely should be considered. In addition, the impact of earthquakes must be taken into account. In fact, the rock mass strength is very difficult to be assessed which causes the difficulty of analysing rock slope stability. Therefore, an empirical failure criterion, the Hoek-Brown failure criterion, has been proposed. It is one of the most widely accepted approaches to estimate rock mass strength. The rock mass disturbance is important and was found having significant influence on evaluating rock slope stability, especially for rock slope with poor quality rock mass. In the Hoek-Brown failure criterion, the disturbance factor can represent the level of the rock mass disturbance which would provide a reasonable basis for estimating rock mass strength.This research will not only discuss the slope factor of safety, but also consider the influence of the seismic force on rock slope stability assessment using pseudo-static method. In practice, only horizontal seismic coefficient is used. Various magnitudes of the disturbance factor and recommended blasting damage zone thickness are also taken into account. The blasting damage zone thickness considered ranges from 0.5 to 2.5 times of slope height. The research results have potential to be extended and then sets of comprehensive stability charts can be provided for the rock slope stability evaluations. They will be convenient tools for practising engineers. In this study, finite element upper bound and lower bound limit analysis methods are employed. Their applicability has been investigated in some previous studies. The differences between upper bound and lower bound solutions are less than ±10% which would provide reasonable and acceptable range for rock slope stability safety factor estimation."

Jan 1, 2015

By Alexander Smith

"The Silver Standard mine is 6 miles by road northwest of New Hazelton, B.C., a small town on the northern line of the Canadian National Railways some l 75 miles northeast of Prince Rupert. The deposits lie on the western slope of Glen Mountain (elevation 2,100 feet), with the main adit, mill, and campsite at 1,300 feet. The deposits were found in 1910, shortly after the area had been swept by forest fire and during a period of prospecting activity spurred by the construction of the railway. Crude ore only was shipped until 1918, when a 50-ton gravity mill was built which operated until 1922. Gross value of production of this period, at present metal prices, was about $1,000,000.In 1947 the present company was formed and a 50-ton flotation mill installed. Production from 1948 to 1955 has been in excess of $10,000,000, and net smelter returns about two-thirds of this amount. More than 60 per cent of the production and 80 per cent of the profit have come from one vein, No. 6. On the 1,300-foot level this vein had an ore section 375 feet long and 2.5 feet wide, averaging 0.18 ounces gold, 85 ounces silver, 10 per cent lead, and 16 per cent zinc.The mine workings follow a simple pattern. A series of parallel veins, striking northeast and dipping southeast, are numbered from 0 to 12, starting at the northwest. They lie within a zone trending south 70 degrees east, and the distance between Nos. 0 and 12 veins is about one mile. Recently several cross-veins, dipping northeast, have been found between Nos. 8 and 12. They are apparently of the same type as the others, though they may not have the same structural controls. They could be an important new source of ore.The mine has been developed principally by crosscut adits at the 1,500- and 1,300-foot levels and a 510-foot shaft with levels at I, 150, 1,000, ·and 850 feet. Drifts and crosscuts total 17,000 feet, chiefly on Nos. I, 4, 6, 7, and 8 veins. The drifts are stoped by shrinkage, top shrinkage rill, cut-and-fill with resuing, or open stope methods. The mill produces a lead-silver-gold-copper concentrate and a zinc concentrate, with indicated recovery of more than 96 per cent (4)."

Jan 1, 1949

By Gerald A. Crawford

"1880s - Unmasking the Sudbury Basin Falconbridge Ltd. turned 60 in 1988 but its roots go back more than a century to the discovery and development of the mineral resources of the Sudbury Basin. Nickel was first identified in the area in 1856 but sustained exploration awaited the early 1880s and the advance of the CPR westward from Sudbury. Rock was laid bare in the right-of-way and by forest fires and logging. Reddish oxidized outcropping or ""gossan"" were exposed, heralding the presence of nickel as the shape of this astounding geological formation gradually began to emerge. Many great mines of the area were opened up in this period on the south rim of the Basin west of Sudbury. To the east the ore dips below overburden, however, and Falconbridge remained a mere dot on the map until the turn of the century when Thomas Edison detected a magnetic occurrence by dip needle."

Jan 1, 1999

By A. Bascetin

The mining industry in Turkey has been developing in the last 10 years. Mine wastes and tailings have thus been increasing, depending on the rising production rates. There are many problems related to accidents that occur in tailings damps. In this study, the applicability of the surface paste disposal (SPD) method for a Pb-Zn mine in western Turkey was examined. Density was 3.084 and specific surface area was 0.899 m2/g. The mill tailings have very low plasticity and show characteristics of clays and silts in terms of the Unified Soil Classification System. The particle size distribution of the tailings was 39.2% particles finer than 20 µm. The results of this study showed that the physical and chemical properties of mill tailings make them suitable for SPD.

Aug 1, 2013

By Leroy G. Hagenbuch

"IntroductionAn integrated Truck Management Information System (Truck MIS) evolves as seemingly separate functions combined to yield a single homogeneous result / product. The concept of a computer integrated Truck MIS includes truck dispatch, truck maintenance management and truck weighing.A truck management system is only as good as the information / data bases input to it. The concept of an integrated system works best with automated data gathering. Automated data gathering provides timely and normally, the most accurate historical information/data bases. Data bases are, however, historical, which is a problem to be overcome in a computerized truck management information system, especially in regards to a truck dispatch system as loading sites / loading conditions and some dump site conditions continually change. As these changes occur, data base s must change as well, and thus the need for the continual update of data bases that the computer uses for a Truck MIS.For a truck dispatch system to achieve reliability / viability, dispatch decisions need to revolve around real time data gathered in an automated way. Integration of all truck management function s through automated data gathering lead s to the best over-all system.An integrated system is a combination of many different factors and is only as good as each factor / building block put in the system. With the proper building blocks, an integrated Truck MIS is very feasible - an MIS System that includes weighing, time and motion study, truck maintenance and maintenance management, as well as truck dispatch.Truck MISAn integrated Truck MIS provides for truck dispatch as the ultimate end result, a truly automated /integrated truck dispatch system (to function non-operator-dependent). The following comprises its central building blocks:-Time.- Haul Cycle Segments.Truck Status, i.e. Load on truck,- being loaded,- loaded,- unloaded,- in service,- out of service.- Vehicle Location.Loading Equipment Status.- in service,- out of service.Optional- distance travelled in each haul segment.With this data provided on a non-operator dependent basis, it is possible to build good data bases so a computer can make truck dispatch decisions. Knowing what information is needed as the central core to a good Truck MIS, how can this information be automatically gathered and fed into a central control computer?"

Jan 1, 1986

By S. Uddin

Microsegregation manifests its effect on microstructural evolution of cast or welded alloys which in turn influences their important mechanical and electro-chemical properties. In the above context, mathematical models of microsegregation have proven to play an important role in estimating the nature and predicting the effects of this phenomenon in wide varieties of solidification processes. Microsegregation models are useful in designing new commercial alloys, in estimating the homogenization time for the post-process heat treatments and in many other practical applications. In this study, two semi-analytical models of microsegregation were developed to predict the concentration field of solute in the liquid and solid regions for dendritic solidification of binary metallic alloys. Both models assume that the growing dendrites are cylindrical in shape. This assumption is more realistic compared to the common assumptions of plate like dendrites that most of the earlier researchers employed in their microsegregation modeling study. The solute redistribution profile, in the developing solid layer, necessary to determine the back diffusion parameter was derived from Fick's second law for the model without coarsening. The application of this parameter in a wide range of conditions and the use of its basic form in the model with coarsening has been verified. The concept of coordinate transformation and enhancement of back-diffusion Fourier number were used in deriving the model which took into account the coarsening of dendrites. The models are then extended to deal with rapid solidification and peritectic transformations and the results were compared with relevant experimental data for Aluminum and other alloys. A good agreement between model predictions and experimental results was found.

Jan 1, 2007

By Trygve Høy

"Important stratabound lead-zinc deposits in southeastern B. C. occur in rocks that range in age from Proterozoic Helikian to Middle Cambrian. The Sullivan deposit, and similar but smaller deposits in the Sullivan area are the only important deposits that occur in clastic rocks with little carbonate content. The highly deformed and metamorphosed Shuswap deposits, of probable Hadrynian to Helikian age, are in mixed marble, calc-silicate gneiss and pelitic schist assemblages, rocks transitional between clastics and carbonates. Deposits in younger rocks, such as the Kootenay Arc deposits in the Lower Cambrian Badshot-Reeves Formation and the Monarch -Kicking Horse deposits in Middle Cambrian dolomite, are hosted by platformal carbonate rocks.An important regional control for the distribution of deposits appears to be a northeast-trending tectonic zone that crosses the Rocky Mountains and the Purcell Mountains just south of the Sullivan area, and projects southwestward toward the Salmo camp at the southern end of the Kootenay Arc. Pronounced northeast-trending gravity and magnetic anomalies, basic volcanism of Hadrynian age (the Irene volcanics west of Creston) and a preferred alignment of Mesozoic granitic rocks indicates that the zone coincides with deep crustal structures. Within the zone are dramatic and abrupt thickness and facies changes in Hadrynian Windermere rocks and lower Paleozoic rocks, and on the east side of the Rocky Mountain Trench in Helikian Purcell Supergroup rocks, indicating that block-faulting was active intermittently within the zone from Helikian through to lower Paleozoic time. Right-lateral and reverse movements occurred on northeast-trending tear faults within the zone in Mesozoic time."

Jan 1, 1982

By A. Zhecheva

A commercial titanium alloy Ti-6Al-2Sn-4Zr-2Mo has been gas nitrided in this work, in pure nitrogen atmosphere in order to improve its surface properties. The nitriding was performed at two different temperatures, 950 and 1050°C, and for different periods of time. The influence of the initial surface roughness on the diffusion process and on the surface properties of this material was studied. The influence of the processing parameters of gas nitriding on the surface roughness, phase modifications and the microstructure of the nitrided layers has been analyzed using atomic force microscopy (AFM), X-ray diffraction and optical microscopy. Microhardness profiles were obtained on cross-sections of the samples after gas nitriding applying a Knoop indenter.

Jan 1, 2004

By K. Kawagishi, J. Ang, A. Sato, H. Harada

"We have recently proposed a new ""EQ Coating"" bond coat system to solve the problem of secondary reaction zone (SRZ) formation caused by bond coats applied on the hot section components of modem jet engines. A viable EQ coating alloy was previously found for a representative znd generation superalloy, TMS-82+. It would be ideal if the same could be done for a representative 5th generation superalloy, TMS-173. In this paper, we will show that the coating with the best oxidation resistance in the EQ system of TMS-173, although possessing considerably better resistance than TMS-173, did not have sufficient oxidation resistance. In order to improve the oxidation resistance of the EQ coating for TMS-173, we have ventured into ""Pseudo-EQ coatings"". A diffusion couple between the Pseudo-EQ coating and TMS-173, after 300hrs at 1100°C, had a 0-15µm IDZ, but no SRZ. A diffusion couple between a conventional coating and TMS-173 formed a 80-150 µm SRZ. After a 1 hr oxidation test at 1100°C, the PseudoEQ coating formed a thinner oxide layer than the EQ coating, but still contained NiO and spinel. Aerospace"

Jan 1, 2006

By Muhammad Zaka Emad

Cemented rockfill (CRF) is a mixture of rock aggregate and binder slurry. It is a stiff backfill, it is fast curing, its operation is simple, placement rate is reasonably high and no drainage is required. The main demerit of CRF is failure of exposed faces causing ore dilution, which can be associated with the fill placement method. This paper discusses the effects of two placement methods being practiced by mines and numerical modelling is used as a tool to evaluate stability of the backfill column. In the first method, practiced in longitudinal retreat mining of a narrow vein deposit, the rockfill is hauled and dumped into the stope with a LHD scooptram. Once dumped, cement slurry is sprayed into the stope. In the second method, a raise situated centrally above the mined stope is used to convey already mixed CRF into the stope. Stopes are mined out transversely in primary and secondary blocks in a sublevel stoping system. Problems of poor mixing in combination with both placement methods are discussed. The two backfill placement methods are analyzed for blast induced vibration damage using the dynamic Flac3D modelling technique accounting for rock mass and fill damping characteristics, and a pressure wave is used to simulate the blast vibrations. Fill heterogeneity has also been considered for the two cases and results are compared.

Jun 1, 2012

By B. J. McKay, K. Schneider, H. B. Nadendla

"A series of experiments were conducted to assess the impact of zinc additions on the morphology and soundness of the interface of Al-Mg compound castings. It was found that high amounts of zinc (>15wt%) are necessary to affect the morphology and microstructure of the interface. At these zinc concentration levels, the formation of the zinc-rich phases t1 and ? were observed along with the Al-Mg intermetallic phases ß and ?. The micro hardness of all phases across the interface was found to be similar, ~200 Hv. This even hardness gradient is thought to avoid mechanical failure in the interface under load. In samples containing large concentrations of zinc (>15wt%), the formation of bubble shaped voids as well as the complete transformation of the aluminium alloy was observed. While the exact mechanism causing the voids has not been identified it thought to be connected to the Kirkendall effect. INTRODUCTION Conventional single material components are increasingly failing to meet the demands, in terms of cost and performance, of the transport industry. Compound castings have been identified as a promising alternative as they possess a number of benefits (Bührig-Polaczek, Röth, Baumeister, Nowack, & Sussmann, 2006). Compound casting is described as a process in which a metallic melt is cast around a solid metal shape. Preliminary research has concluded that for the stability of such a resultant hybrid structure it is imperative that both metals form a sound connection via form closure, force closure, or material bonding. A continuous material bond between the two metals is also highly desirable as the formation of a crevice between the metals could lead to galvanic corrosion, particular in a humid environment (Lao, 2013; Rübner, Günzl, Körner, & Singer, 2011). Even though Al-Mg compound castings could offer the biggest potential in terms of weight savings to date only a limited amount of research concerning such structures exists."

Jan 1, 2018

By C. Guo, R. J. Chalaturnyk, M. MacKinnon, J. D. Scott

The Mildred Lake Settling Basin (MLSB) is the largest disposal site for mature fine tailings (MFT) at the Syncrude Canada Ltd. oil sands plant. Over the past years (since 1996), there has been a marked change in the densification behaviour of MFT in the MLSB. Methane-producing microorganisms, known as methanogens, have become very active, and large amounts of biogas (mainly methane) have been produced. Continued field monitoring of the MLSB has provided convincing evidence of the rapid densification process (rapid water drainage from the tailings) at the area with intense microbial activity. This rapid densification has caused pumping challenges in the transfer of fine tailings from the Mildred Lake Settling Basin for the creation of composite tailings. It may also have potential positive effects in accelerating the reclamation of the oil sands fine tailings. A field and laboratory research program was carried out to study the mechanism leading to the rapid densification phenomenon. A review and discussion of the research program is given and some results from the field investigations and small-scale column tests are presented in this paper.

Jan 1, 2007

By M. Sugamata

Creep tests were carried out for friction welded joints of AZ31 magnesium alloy rods under an ambient atmosphere at temperatures between 423 and 573K. Tensile tests were also performed at temperature from room to 573K. The obtained results were compared with those of the base metal. AZ31 extruded rods of 19 mm in diameter were joined by brake type friction welding machine. Creep and tensile specimens were machined out of the welded rods. The tensile strength of the joints was almost the same as that of the base metal at room temperature. However, joint efficiencies of welded joints were about 92% at above 373K. The elongation of welded joints was lower than those of base metal. The welded joints showed appreciably higher minimum creep rate and shorter rupture time than the base metal at all the tested temperatures and initial creep stresses. Creep and tensile rupture of the joints always occurred at near the weld interface. Thus, friction welded joints of AZ31 alloy showed lower creep and tensile strength than the base metal at elevated temperature.

Jan 1, 2006

By Narasimha Mangadoddy, Balraju Vadlakonda

"Bubble size distribution, bubble rise velocity and gas and solids hold-up are the key major hydrodynamic parameters that are extensively used for column flotation design and performance evaluation. The distribution characteristics of mean gas hold-up is studied using electrical resistance tomography (ERT) coupled with pressure transducers (PT) in column flotation. The ERT technique facilitates non-invasive and nonintrusive visualization of different flow regimes even in the presence of particles in a column flotation. The effect of superficial gas velocity, feed flow rate and slurry height in the column on mean gas hold-up and its radial distribution has been analysed for both two-phase and three phase flows. Experimental results show ERT coupled with PT is suitable as an online monitoring tool to provide useful information on the three-phase flow hydrodynamic parameters of the column flotation. The tomography images, which were generated using a modified sensitivity back projection algorithm, are employed to explore the influence of parameters in two phase flow in column flotation. Mean gas and solids hold-up values extracted from ERT have been critically analysed for the column operating in various flow regimes. Gas hold-ups and solid hold-ups results using ERT are in very good agreement with conventional estimation and correlations obtained using pressure transmitter methods. Meanwhile, the results show that the gas hold-ups in the centre region increase constantly with an increase in the superficial gas velocity, whereas solid hold-ups distribution is very heterogeneous for high gas velocity. Furthermore, the results also indicate that ERT is a very powerful tool for diagnosing the 'inside' flow behaviour of gas –liquid–solid three phase flow in columns operating for a wide range of flow regimes.INTRODUCTIONColumn flotation is perhaps the most significant technological innovation in flotation machinery and its successful introduction in mineral processing plants. There has been a lot of attention in understanding the various mechanisms that take place during operation. Due to many reasons column flotation has been found to yield better performance than conventional flotation cells, particularly with fine particles (Honaker & Mohanty, 1996). Due to complex flow behavior in nature, the performance of column flotation influenced by on flow characteristics of the process, (Jena, Biswal, Das, & Reddy, 2008). Previously basic research is carried out to evaluate the effect of the hydrodynamic parameters on performance of column flotation (Finch, J.A.;Dobby, 1991). Many techniques have been developed that could be used to experimentally analyse the hydrodynamics of column flotation. Tomography techniques are well sound to investigate the hydrodynamics for process equipment. Among that, Electrical resistance tomography (ERT) has been used in various industrial investigations for visualization of the concentration profiles and characteristics of the fluid dynamics in gas–liquid two-phase systems (Wang, Dickin, & Mann, 1999; Williams, R.A; Wang, 2000). ERT is an imaging technique to analyse the internal flow behavior non-intrusively. ERT provide very useful tools for measuring and monitoring of column flotation process on line. Due to its high speed capability, low cost measurement, robust sensors, ERT is considered to be the most powerful tool among other tomography techniques such as positron emission tomography (PET), magnetic resonance imaging (MRI), optical and infrared tomography. Application of ERT has extended in process systems due to its improved speed of data acquisition, sensitivity, flexibility and noise immunity. In multiphase flow of bubble columns, gas hold-up and solid hold-up are explored by using the coupling of ERT and difference pressure method (Jin, Han, Yang, & He, 2010)."

Jan 1, 2016

By R. M. T. Raja Ismail

Shipping and stevedoring industry is going to witness a massive increase in the amount of containers to be handled while land constraints become more critical. Offshore transfer operations hence offer a preferable solution to deal with the surge in cargos rather than to expand the port outwards. Recently, there has been increasing research interests on offshore crane automation. Suspended cargos in a ship-mounted crane system are caused to swing due to the vibratory motion of the ship induced by ocean waves, which can lead to collision between cargos and deck. Therefore, it is vital for offshore crane systems to satisfy rigorous requirements in terms of safety and efficiency. This paper presents the modelling and control development for offshore boom crane systems. A second-order sliding mode control law is proposed for trajectory tracking and sway suppression control, making use of its capability of actuator chattering alleviation while achieving high tracking performance and preserving strong robustness. The asymptotic stability of the closed-loop system is guaranteed in the Lyapunov sense. Simulation results indicate that the proposed controller can significantly reduce the effect of disturbances coming from gusty waves and other dynamic loadings.

Aug 1, 2013

By Maureen A. McGuinness

Kidd Mine, an Xstrata Copper operation in Timmins, Ontario, operates a 400 t/h pastefill plant that provides paste backfill to stopes at the depth of 9000 ft. This paper will cover the operating practices of the pastefill plant both on surface and underground, highlighting tailings excavation, paste mix design and the pipe maintenance program. Kidd Mine, une opération d'Xstrata Copper à Timmins, Ontario, opère une usine de remblai d'une capacité de 400 t/h. Le réseau de distribution permet d'acheminer le remblai en pâte juste qu'à une profondeur de 9000 pi. Cet article couvrira le mode opératoire de l'usine de remblai tant à la surface que sous-terre en mettant l'emphase sur l'extraction des rejets miniers, la recette du remblai en pâte ainsi que le programme d'entretient du réseau de distribution.

Feb 1, 2008

By S. Isaac

The networks of interdependencies between elements in building projects, referred to as the project topology, are often complex and difficult for project teams to comprehend. It is hypothesized that a graphbased model of the project topology can be based on existing data in Building Information Models (BIM), facilitating the application of graph-theoretic algorithms to analyze this topology. The objective of this study is to propose an outline for the synthesis of BIM and graph theory. This synthesis is expected to contribute to the development of new tools for representing, managing and analyzing the design of building projects.

Aug 1, 2013

By K. Esmaieli

The Synthetic Rock Mass (SRM) approach has the potential to capture the mechanical behavior of large rock samples. In order for the method to be used with confidence it is necessary to gain a good understanding of the factors that influences its analysis. This paper reports on a series of numerical experiments, based on field data from a large underground mine, to investigate the influence of intact rock mass strength of a constructed (SRM). A jointed rock mass defined by three fracture sets was simulated by a Discrete Fracture Network (DFN) model. The generated fracture network was subsequently embedded into six different intact rock samples, simulated by a bonded particle model in PFC3D, generating six SRM samples. Each intact rock sample was 7 m × 7 m × 14 m and assigned different mechanical properties. The first intact rock sample was assigned a uniaxial compressive strength of 205 MPa and an elastic modulus of 104 GPa with the UCS and elastic modulus values of the other samples downgraded by 10% until the weakest sample had a UCS value of 121 MPa and an elastic modulus value of 61 GPa. All SRM samples were loaded under uniaxial compression to obtain the complete stress strain curve. It was observed that the compressive strength of the rock mass samples decreased by 10% following the same trend as the rock material strength. However, the elastic modulus of the rock mass samples decreased non-linearly with reduction of the elastic modulus of the rock material. The Poisson's ratio of rock mass samples does not change with decrease of rock material properties. For the post-peak behaviour the results show that both the brittleness index and the residual strength of rock mass samples decreased as the rock material properties degrades.

Aug 1, 2013

By J. Zou

Open-cut method is most widely used for shallow utility lines installation. In urban areas, unexpected utility lines buried underground potentially turn open-cut construction into a highly risky operation, increasing the construction cost while presenting additional safety hazards. Examples include: breaking existing water main leads to flooding the downtown area; hitting an unexpected gas line causes an explosion. To prevent those accidents, the current practice is to stake out the underground utility lines before the open-cut construction based on the as-built information collected from various utility companies and government agencies. However, the as-built information is not always complete and accurate. To verify the locations of some important utility lines and protect them against damages caused by open-cut construction, new techniques and technologies are used but have their limitations in revealing underground utility lines, such as hydro vacuum method and ground penetration radars (GPR). The emerging technology of ultra-wide band (UWB) radar holds the potential to provide a cost-effective, non-destructive detection method. In this paper, a critical review of current practices and established methodologies is given. The functionality and working mechanism of UWB radar technologies are described. Application potential in open-cut construction is demonstrated by conducting lab experiments. Testing set-ups for detecting unexpected utility lines in soil along with preliminary results are presented. The effects of soil moisture content on the detection range are discussed. Research findings are summarized and constraints are discussed in conclusions.

Aug 1, 2013