Search Documents

Jan 1, 1920

By D. G. Brubaker

IN the history of geophysical exploration by the electromagnetic method many procedures and types of equipment have been used. Source arrangements for surveying on the ground have included long wires grounded at both ends, large loops laid on the ground, small horizontal loops, and vertical loops of moderate to small sizes. Receivers have ranged from simple direction-finding coils to combinations of coils and a bridge circuit to measure quantities related to both amplitude and phase of the received signal. Primary considerations in choosing a survey method are effectiveness, speed, and economy. In work on the ground in which locating and out- lining conducting bodies is the objective, methods using a vertical coil and a simple direction-finder have been found the most satisfactory. Mason1 and his associates applied such a method by setting up a vertical source loop, fed with low frequency alternating current by an engine-generator set. With a direction-finding receiver they surveyed lines along the ground in the plane of the loop, taking measurements at each station of the azimuth and inclination of the magnetic field direction. Deviations of this direction from the normal direction, that is, from horizontal and perpendicular to the traverse line, indicated a conducting body. Recent advances in design of light-weight generating equipment have encouraged using this and similar methods.

Jan 7, 1957

By Floyd C. Kelley

The literature of the last decade is rich with information relating to the cause and means of control of grain growth in pure metals, but is deficient concerning the role diffusion plays in grain growth, although during this period many new products that depend on diffusion for their properties and success have been developed. Among these are cementation products, produced by calorizing, chromizing, and carbonizing, or by mixing, pressing, and sintering powdered metals such as Syke's iron-molybdenum die material, Genelite, and Elcon Metal (copper, tungsten) etc. This is a fertile field for new developments, but before it can be utilized to the fullest extent there must be a thorough invest,igation of the processes by which diffusion and grain growth take place. Grain growth plays no small part in the process and certainly the physical properties of these products depend largely on the grain size and state of equilibtiuni produced. The field of investigation does not end here, but may be extended to the effect of diffusion on impurities, segregations and dendrites in the broad field of ferrous and non-ferrous metallurgy. A number of years ago the author investigated several of these cases and noted, with striking regularity, the pronounced effect of diffusion in promoting recrystallization and grain growth. An example of this phenomenon is given in his paper on chromizingl which contained photomicrographs of large radial or columnar crystals produced by diffusion of chromium in the solid state. Another manifestation of this effect was discussed by Austen,2 who showed the part hydrogen plays in the decarburization of steel and the effect of diffusion upon the grain structure. Diffusion of Metals in Iron It seems to be a general law, whenever two metals are brought into intimate contact at a temperature at which diffusion takes place, that a comparatively rapid, or abnormal, grain growth results. The grains

By Carl S. Westerberg

Coal preparation practice and trends follow, among other factors, production trends in any given area. Considering an area the size of a state, some broad predictions may be made after a review of the annual production records. As this paper is concerned with coal cleaning in Utah, data on its production record are pertinent. Table 1 shows the percentage relation of coal production in Utah to the national total as a single figure up to 1935 which includes all recorded production. From 1935 through 1947 these data are presented by years. It is evident from these data that the production of Utah coal is ascending. The reason for this trend is industrial growth in the West resulting in a population shift which accounts for more people than the normal increase would supply together with Utah's relatively fortunate geographical location and its abundance of good-quality coal. Utah is also hopeful of supplying more coal in the future for West Coast power generation.I Utah is not unique among coal-producing states in anticipating greater production in the future. Coal Age states, " . . . it is now evident that a billion-ton yearly production of anthracite and bituminous by 1960 is not impossible." At present it is about 610,000,000 tons, the all-time high in 1947 was approximately 687,803,000 tons. In view of the trend outlined above, obviously more new mines will be required to produce the necessary tonnage. These in turn will require more mechanical cleaning capacity. Whether the required new plants will be built by present operators of mines and mechanical cleaning plants or by others now in the business of producing coal or expecting to enter it is of no consequence in this discussion. The fact is— they will be built.

Jan 1, 1950

By M. S. Azun

I do not at all agree with the basic points of the author's conclusion. The use of lognormal or normal model to respond to the attribute distribution function should be carefully questioned. If fitting a distribution function of regionalized variables is the main purpose, one may employ one of certain well-known families of distributions, such as Tukey's lambda functions. For instance, the distribution function of gold assay values studied by Krige (Krige, 1978) may fit the lognormal model well, but one should not exclude that the basic idea of using the distribution function of observed data is to have insight of the physical mechanism that gives rise to the regionalized variables. In Applied Geostatistics, there are many examples considering only lognormal or normal model for the distribution function of geostatistical data even if the observed distribution function has more than one mode. Of course, using lognormal or normal model always brings easy computation for the statistical properties of attribute under investigation. Still, I object to adopting any of those models without having any "statistical inference." As seen from the author's paper, lognormal and normal model produce the same average grade (0.746% copper) and almost the same sample variance. Therefore, which model explains the probabilistic behavior of ore deposit is always an important question that should be replied by the geostatistician using not the common sense but the statistical and probabilistic methods. For kriging procedures, such as linear kriging, one should describe either second order stationary properties or bivariate properties of regionalized variables. Second order properties, such as correlogram function or semivariogram function, are the inputs of the so-called kriging equations. Selecting one of any useable models for the observed second order properties, such as spherical model, is not easy since the models are not based on the correlations structure of the regionalized variables. How those models being used in Applied Geostatistics can be distinguished is another important problem. One may develop many fitting models to respond to the observed second order properties of regionalized variables. However, I suggest that the model should be based on the probabilistic behavior of geologic process. The author used spherical model for copper, wolfram, and silver grades from isotropic sector (Fig. 2, 4, 3b) and exponential model for silver grades obtained along raises in isotropic sector (Fig. 3a). It seems that those semivariograms given in Fig. 3b for isotropic sector and Fig. 4 may be described by the random model implying that regionalized variables do belong to renewal process (Azun, 1983). If I knew the total number of data points used in estimation, may prove that my conclusion is correct. Recalling the test for the first order sample correlogram value, the test procedure is introduced such that the independence is rejected if [Ir(1) I > Za/2 N1/2] where Za/2 is the a/2 percentile of standard normal random variable and N is the total number of samples used in estimation (Azun, 1983). For the other observed semivariogram functions I suggest the author might try to use the so-called "Markovian model" describing not only the correlation structure of regionalized variables, but also the physical mechanism that produces the regionalized variables. The Markovian model for correlogram function is, [p (h) = T 1ih , h>1 ,] The Markovian model for the other second order properties of ReV are also derived. [T veß], in the above equation, show, for example, structural change and mineralizational variation in a considered deposit, respectively (Azun, 1983). In selecting any model, it is not easy to search for the "best" response to the observed second order properties of ReV's. The Markovian model, based on a theoretical understanding of underlying mechanism, gives more information about the occurrence of regionalized variables and respond to all properties. Random model and the so-called hole effect structure can be easily defined as a special form of the Markovian model (Azun, 1983). An estimation of any second order properties of regionalized variables may be computed through (N-1) lag. However, the dependency between the regionalized variables may be deemed to have "died out" after the so-called stationarity width (range). In practice, the estimation is carried out through 25% of the total number samples available. Therefore, the large fluctuations around the zero level and variance for correlogram or covariogram function and semivariogram function, respectively, cannot be observed. Since the number of pairs involved in the estimation of higher second order properties is small, the estimation variance at those lags is large. There is no need to show higher order estimated second order properties. After using one of any kriging procedure for block estimation, the distribution function of average grades has the same mean as the drilling sample grades but

Jan 1, 1986

By L. C. Glenn

In the preparation of this paper the writer has drawn freely upon the writings of Orton, Munn, Shaw, Mather, Miller, Hoeing, St. Clair, Jillson, and others, as well as upon his own personal knowledge of the fields of both states. It is to be regretted that certain data gathered by him and his assistants last fall are not available for publication. Oil in Tennessee A few wells drilled for brine for salt making in Tennessee between 1820 and 1840 obtained oil, but no definite search was made for oil until just after the close of the Civil War. Active drilling was then begun in Overton and counties southwest of it on the eastern half of the Highland Rim. A number of strikes were made at shallow depths in the basal part of the Mississippian but the wells were soon exhausted and abandoned. Drilling was revived, about 1892, when the Spurrier district in Pickett County was developed and was followed by the Riverton district in the same county in 1896. A pipe line was laid from the Wayne County, Ky., fields and about 60,000 bbl. of oil were run before a very heavy slump in the production, a failure to find an extension of the field, and excessive local taxation caused the removal of the pipe line in 1906. There was then no production in Tennessee until the discovery, in 1915, of oil near Oneida, Scott Co., at about 950 ft. (289 m.) in fissures in the Newman or St. Louis limestone. This field, however, soon failed and was abandoned. In 1916, oil was found at Glen Mary, Scott Co., in the Newman limestone at a depth of 1232 ft. (375 m.) A number of wells have since been drilled there, some of which were dry while others, close by, were produccrs. The largest one yielded, at first, about 340 bbl. per day and produced for several months, when it suddenly went dry. Several of the first wells began at 6 or 8 bbl. per day and are still maintaining that output. Production is from a fissured part of the limestone and varies greatly in accordance with the size and extent of the ramification of the

Jan 1, 1921

By J. L. Huitt

Some of the problems associated with the acid treatment of oil wells are related to the control of the composition of formation treating acid. One test usually made at the well site is the determination of the acid concentration. Heretofore, the concentration of the acid delivered to the well site usually has been interpreted from a hydrometer measurement of the specific gravity of the acid solution. Experience has shown that this indirect measurement of acid concentration can give erroneous results. More reliable results are, of course, obtainable by the titration method; however, this method is not easily adapted to well-site testing. To overcome these objections, a device was developed for use in the field which permits determination of the concentration of treating acid both quickly and accurately at the time of the acid treatment. Its purpose was three-fold, namely, to determine: (1) the concentration of the fresh treating acid, (2) the concentration of the spent acid returning from the well, and (3) the solubility of formation samples in treating acid. The device was designed to operate on the principle of the stoichiometric relation between the calcium carbonate—hydrochloric acid reaction and the carbon dioxide gas resulting from the reaction. By confining the carbon dioxide gas within a reaction cell of the type shown in Fig. 1, the pressure inside the cell is increased. Since the pressure is an indication of the quantity of carbon dioxide gas produced, it may be used to estimate the concentration of the acid, provided an excess of calcium carbonate is used. From information on the volume of the reaction cell, the volume of acid used, and the pressure and temperature of the gas after the completion of the reaction, the concentration of the treating acid may be computed. The reaction cell is essentially a pressure vessel equipped with a pressure gage, temperature indicator. pressure release valve, and a special sample holder. Plastic and metal construction makes the cell both rugged and acid-resistant. The sample holder is so designed that a sample of calcium carbonate (or sample of carbonate rock) can be dropped into the acid

Jan 1, 1956

By JAMES J. DOUGHERTY

THIS paper is a general discussion of developments in our reverberatory smelting practice during the past five years. It deals briefly kith changes in furnace types ; changes in furnace feed ; de- crease in the amount of slag produced; relation of slag assays to copper loss in the slag; increase in the grade of matte and furnace capacities. The reverberatory is but one unit in the copper plant ancl 'any discussion of it must be considered with reference to the other units and to the plant as a whole. A change in the grade of concentrates at the concentrator puts an entirely new aspect on reverberatory smelting and, likewise, a change in the grade of matte at the reverberatory affects the rate and cost of converting. The introduction of differential flotation in 1927 had a very marked influence on all smelting operations and its effect will be noted

Jan 1, 1930

By Reich G. Gaertner

Almost every mining engineer who has had charge of extensive underground workings will have observed how often directions as to course and levels, deduced from careful theodolite measurements, have been slighted or disregarded by mine-foremen and shift-bosses, and how often these tedious measurements had to be repeated, even for minor works, such as ventilation-drifts or winzes, in order to carry them on with the required or desirable regularity. Besides the time and care absorbed by theodolite surveys, there is still another inconvenience, namely, the interruption of transportation during the time of observation, and, in heavily producing mines, the loss consequent upon such interruption. In proposing a subsidiary instrument to the theodolite, it is my purpose not to present a new one, but simply to recommend to the consideration of American mining engineers one which has been in use in nearly all the prominent mines of Europe for many decades, and has there fulfilled its purpose admirably. I refer to the mining or hanging compass (Fig. 1). At my urgent request, the firm of Keuffel & Esser, of New York, have lately constructed this instrument in a form which adapts it better to use in this country. By the addition of the trigonometer, the banging compass becomes at once an instrument which will prove of value even in the hands of foremen and practical miners, who have little or no knowledge of trigonometrical calculations. Probably many will say that me have no use for needle-instruments ; that our modern theodolites are of such perfect description, and leave so little to be desired, that the introduction of a compass would look like a step backwards compared to the fine results obtained by angle-instruments. I concede all that, and do not pre-

Jan 1, 1886

By W. L. Maxson

GRINDING is one of the major problems in present-day milling practice, and in many cases, it represents one of the main items of expense. It becomes necessary from time to time, to compare grinding practice in one plant with that in another plant, and in making such comparisons, it would be very desirable were there available accepted standards of grinding resistance for the materials involved. Such a set of grinding constants would also be of considerable value in planning new grind-ing installations. There have been some tests published, notably those of Lennox,1 which have been generally cited as a basis for such comparisons. Most of the tests made have been based upon open-circuit grinding, in which a weighed sample of the ore was crushed to pass a certain mesh, and was .then ground under certain conditions, either wet or dry. Grinding was continued until a certain desired fineness was attained, and the compari-sons based on the grinding time, or the samples were all ground for a certain specified length of time and comparison based upon the screen analysis of the products. Such methods have been acceptable where the ore was to be ground in open circuit. However, in modern milling practice, many grinding problems involve closed circuits wherein the oversize returns to the mill and a large circulating, load builds up. A method of testing that does not take the circulating load into consideration might yield results quite different from those to he expected in practice, because the minerals in the ore that offer maximum resistance to grinding will accumulate in the circulating load up to an equilibrium value. Thus the nature of the ore charge in the mill at any instant may differ greatly from that of the new feed, with a consequent error in the rate of grinding determined by such a method.

Jan 1, 1933

By H. H. Pentz

THIS paper gives the derivation of a practical formula for the calcu-lation of slopes in seismic reflection prospecting. The derived formula is an approximation and can be used where the slope of the reflecting horizon is less than 250 ft. per 1000 and the depth to the reflecting horizon is greater than 2500 ft. The slope of sediments in oil regions is generally considerably less than 250 ft. per 1000. This is particularly true of the Gulf Coast, with the exception of the salt domes, which usually have the sediments at a greater slope than 250 ft. per 1000.

Jan 1, 1937

By John J. Stokes

Literature on the electrolysis of aluminum from cryolite melts and on the structure of these .melts is surveyed critically. Data on density, freezing point, and other properties are reviewed. Theories of the electrolysis are examined in the light of these data. Two theories are presented which account equally well for the observations. In the early days, the process was thought of as a simple one. The actual mechanisms must be tremendously more complex. This paper is devoted to a rather narrow part of the whole subject, namely, what ions are present in the bath, and by what mechanism the current is carried. Using cryoscopic data, Rolin has determined the following composition for molten cryolite. Na2,AlF 3Na + 3F + AlF [1] He postulates the following equation for the ioniza-tion of alumina at low concentrations Al2O3 Al + AlO2 +O. [2] At concentrations above 0.025M he supposes the ionization of alumina to proceed as follows Al2O3 AlO + AlO2. [3] Rolin's mechanism of electrolysis for the more concentrated solution in Eq. 3 is as follows. The AlO' ion migrates to the cathode. where it is dis- charged, giving metallic aluminum and 0' ions which migrate to the anode. The AlO2- ion migrates to the anode, where it is discharged, giving Al ions and oxygen which reacts with carbon to form CO2 In the case of the extremely dilute solution in Eq. 2 the Al ion migrates to the cathode and the AlO2-and 0 migrate to the anode as before. Another possible mechanism for the electrolysis follows. Ionization of Cryolite Na3AlF4 = 2Na + 2F + NaAlF, Na + AlF-AlF+Na + F [4] Addition of Alumina 2Na2AlF+ 2AL2O4 6Na + 3A12O3F4 [5a] Al2O2F 2AlOF2 [5b] Cathode Process AlF, + 3e+ 4F + Al. [6] Anode Process AlOF2 + 2F AlF1 + 1/20, + 2e or A12O2F + 4F-2A1F; + 0, + 4e. [7] Although both Rolin's and the present mechanism of the Hall process may be qualitatively satisfactory, present data are not adequate for quantitative tests of the theories.

Jan 1, 1959

By Atsushi Watanabe

Japanese Operations Research Committee for Mining Industry is one of the committees in The Union of Japanese Scientists and Engineers. Members of the committee are Dowa Mining, Furukawa Mining, Mitsubishi Metal Mining, Mitsui Mining and Smelting, Nippon Mining and Sumitomo Metal Mining. This paper is an annotated summary of study reports of the committee.

Jan 1, 1969

By H. J. Hager, P. H. Haskell

Fox the past 20 years Alabama has probably led all other coal-producing districts in the proportion of coal prepared by wet washing. All rail mines, with one exception, and a high percentage of truck mines now wash all coal under 3 or 4 in. Most of this coal has been cleaned by jigs, though within the past few years one or more wettable plants have been installed for cleaning some of the coal from practically all the major seams. These tables supplement either single or multicell jigs usually for one of the following purposes: (I) to wash the finer sizes of raw coal, (2) to rewash the finer sizes of washed coal after they leave the jigs, (3) to retreat jig middling products. The main purpose of this paper is to present cleaning data and operating factors involved in the preparation by tables of the finer sizes of raw coal. The specific information and data pertain to the Mary Lee seam of coal as extracted at three of the mines of the Alabama By-Products Corporation, which produces coal for both by-product coke-oven consumption and the general steam market. The Mary Lee seam has long been recognized by coal-preparation authorities as presenting a very difficult ash-reduction problem. Washing characteristics and physical structure of the seam vary widely. Its thickness ranges from 32 in. with no rock parting to l00 in. with from one to three partings. Over the entire field the bed contains bands of bone and rash. These are distributed throughout the seam cross section. A portion of this bone or bony coal occurs as definite bands 1 in. or more thick; however, some bone is interlocked as layers within the coal and consequently breaks into flakes when the coal is broken. This material is relatively high in ash and its specific gravity varies widely. It has been observed that in most cases in the washing of Mary Lee coal from 10 to 18 per cent of the raw coal is within ±0.10 specific gravity of the attempted point of separation. Therefore with the average wet jig, only with extreme difficulty can a low-ash product be obtained without loss of much marketable coal in the refuse. Without entering into the controversial question as to whether modern washing equipment can efficiently handle simultaneously all sizes up to a given maximum, suffice it to state that the comparatively old-type jigs in service at the Alabama By-Products Corporation's mines could not clean the fine sizes satisfactorily. Therefore, in 1931 that company installed six diagonal-deck tables at the Barney mine, which were installed as an addition to a complete plant cleaning all coal below minus 3 in. The primary washed product is shipped as steam coal and another portion may be prepared as washed nut for domestic trade. These tables are washing 9 tons per hour of raw coal sized to ¼ in. to 0. The washed table coal is mixed with the 3 to ¼ -in. jig product to constitute a 3 in. to o steam coal.

Jan 1, 1941

By Roscoe C. Clark, Henry F. Coffer, J. J. Reynolds

A cement slurry lightweight additive has been adapted in the Conoco laboratories for use in oil well cements. This additive makes possible the use of air to lighten oil well cement slurries. Specifically, the additive consists of small clay bubbles of air having sufficient strength to resist crushing by bottom hole pressures. Clay bubble cement slurry weights can be safely reduced to 10 Ibs/gal. In addition, the clay bubbles tend to act as lost circulation materials and to provide a cement texture which should produce good perforating characteristics. Clay bubble cement is relatively impermeable. INTRODUCTION In many areas in Oklahoma, Texas, Wyoming, and California, difficulties are experienced during casing cementing operations with lost returns caused by the high density of the cement slurries. Wells in these areas are ordinarily drilled with 9 to 11 lbs/gal. drilling mud while neat cement slurries may range in densities from 14.5 to 17 lbs/gal. In these areas, cement slurry weights are at present being successfully reduced to around 12 lbs/gal. (under bottom-hole conditions) by the use of additives in the cement. Since the use of additives will decrease the strength which the cement will develop, the amount of additives which can be used must be based on the minimum cement strength which is acceptable; and thus the maximum slurry weight can be reduced. At present, the minimum tensile strength generally acceptable is 50 lbs/sq in. cured for 24 hours under bottom-hole temperature. This figure is based on laboratory tests1 and on a large number of cement jobs which have been conducted using cement which will not develop more than 50 psi tensile strength in 24 hours under bottom-hole conditions. Two additives are now in use for decreasing cement slurry weights. These are expanded perlite and/or bentonite. The density of both of these materials is greater than that of the cement slurry; however, large quantities of water may be added with these materials without producing a slurry which will settle out of suspension. It is these large quantities of water which contribute to the reduction in the slurry weight. The minimum slurry weight at present obtainable with expanded perlite and/or bentonite is approximately 12 lbs/gal. using a standard Portland cement slurry. In many areas, this weight reduction is insufficient; and returns are still lost during casing cementing operations. In many cases, the pay formation itself is fractured; and a considerable quantity of this cement is lost to the pay zone. It is therefore believed that if cement slurry weights can be reduced to as low as 10 lbs/gal., two advantages will accrue: 1. The lighter densities (10 to 12 lbs/gal.) should mitigate or eliminate lost returns during casing cementing operations. 2. Cement slurries above 12 Ibs/gal. (the present minimum limit ) should develop considerably more strength than is now being obtained. A tensile strength of 100 psi is desirable.' It is the object of this paper to describe a cement slurry having a weight of 10 lbs/gal. and capable of

Jan 1, 1955

By John A. Riddle

Some of the pitfalls and problems in selecting and applying corrosion-resistant materials in uranium plants are discussed, together with some suggested solutions. Includes some cost data.

Jan 1, 1982

"Quiet, please!" is the newest directive being thrust at industry by guardians of the environment-with good reason. In countless cases, industrial noise is literally deafening its listeners, and sooner or later the volume must be turned down. Noise control is already mandatory for many companies under a Federal regulation that went on the books when the Walsh-Healey Act was revised last May 20. The new rules impose specific noise limits on any plants supplying the Government with goods under a contract valued at more than $10,000.

Jan 1, 1970

John Langton, New York, N. Y. (communication to the Secretary*):—In Mr. Firmstone's discussion (Trans., xli., 912) the formula he gives for the periodical payment—his equation (2)—may be simplified in form. The expression given by Mr. Firmstone is : q=S(l+ r) 1-(1+r)-(1+r) r which is a simpler form to use in calculations. Where S= principal to bc extinguished. Q = periodical payment or charge for sinking-fund, r = rate of interest per period (i. e., interest on $1 per period). n = number of payments or periods. Kent's Mechanical Enyineer's Pocket Book, under the heading Annuities, equation No. 5, gives : " The annuity which $1 will purchase for any number of ytlars n is * Received Jiily 16, 1911.

Jan 1, 1912

By George E. Aiken

Modern high speed bucket wheel excavators (BWEs) can dig materials which power shovels can't handle without blasting. Yet these machines are not used in any large, open pit ore mines in the United States. American mining has been slow to use the wheel's special capabilities of high production volume at low cost, due perhaps to a traditional familiarity with shovels and draglines and/or the lack of data of BWE operation, design and cost. Of course, shovels and draglines can not be replaced by BWEs in all cases, but under suitable conditions bucket wheels can out-perform other conventional excavators and simultaneously permit significant operating savings. The latter aspect will become increasingly important to American mining men because as more low to medium grade orebodies are found in regions of thick overburden, better and cheaper ways of uncovering them must be developed. Typical of such areas are the thick alluvial fans of the Southwest and the glacial drift areas in northern regions. Much of this material would handle well with a wheel excavator without blasting, although ripping might be necessary in some cases.

Jan 1, 1966

JAMES E. HOWARD,* Washington, D. C. (written discussion?). -Mr. Comstock, in his classification of the possible causes of transverse fissures, has not quite caught the point of view of the writer. It appears, following the trend of thought in Mr. Comstock's paper, that three groups instead of two might bee enumerated, namely: Fatigue of the steel, quality or, preferably, grade of steel, and mill practice. All steels are liable to failure by fatigue. Stresses that are capable of causing fatigue fractures are modified in their magnitude by the grade of steel acted upon. Mill practice is a term having a very wide meaning and in its relations to the development of transverse fissures has been used with such vagueness as to include many known factors and some that are unknown in the arts. According to the grade of steel employed, fatigue fractures may be accomplished by loads of greater or less magnitude, hence there is a relation between quality, or grade, of steel and the display of fatigue fractures. Mill practice of such an order as to affect the formation of transverse fissures must be shown to exert an influence on the ability of the steel to endure repeated stresses. Different grades of steel are selected for different purposes. Steel employed for fire-box purposes is of different grade from that used for cutting tools; although each grade is suitable for its purpose, and therefore of good quality. It is conceivable that a demand might be made for steel of physical properties in excess of those of any known grade. Under such a demand, impossible of' fulfillment, it would not be proper to refer to the steels as being deficient in quality; to do so would be a failure to recognize that all steels have their limitations in strength. Service conditions peculiar to rails admit of successive advance until the conditions can no longer-be met. The failure of a large number of rails from one type of failure, when all known means of averting the failure have been employed without success, is sufficient evidence that the end is close at hand. Iron rails, for a time, successfully met the requirements of railway service but weight of equipment, motive power, and rolling stock were increased, necessitating the use of steel. The early steel rails enjoyed the reputation of being of excellent quality, although: neither chemically nor structurally would they be so regarded ,lit the present time. They did good service in their day and under the conditions then prevailing were good rails. To meet the increased weight of rolling stock there has been

Jan 3, 1919