Search Documents

By J. C. Bayles

Jan 1, 1885

By J. M. Sands

Describes important developments in, four counties, all of which brought in 40° oil. Indications are favorable for the future, although the daily production of the agate decreased 19,000 bbl. during the year 1923. THE state of Kansas had a daily average crude-oil production of about 87,000 bbl. at the beginning of the year. This showed a steady but consistent decline during the 12 months, so that the daily average production at the end of the year was slightly over 68,000 barrels. GREENWOOD COUNTY THE most active development in Kansas during the year was in Green-. wood County, in a line from near Sallyards in T 25-R 9 to east of Madison in T 22-R, 11. The main part of. this development has been in a succession of pools beginning at the south with the Polhamus pool in 34-24-9 and 3 and 4-25-9; Thrall pool in 28, 29, 32 and 33-23-10; Burkett pool in. 13, 23, 24 and 26-23-10; Seeley and Wick pool in 22, 27, 28, 32 and 33-22-11 and 5, 6, 7, and 8-23-11; Frohm pool in 11, 14 and 15-22-11; Madison pool in 31 and 32-21-12 and 6, 22, 12 and 1-22-11. Some of these pools, have two or more sands, locally called Bartlesville and Burgess Sands. They range in. depth from 1850 ft. near Madison to 2450 ft. near Sallyards, the general dip being to the west. The thickness of these sands runs from 20 to 100 ft., the initial production of the wells being 50 to 1000 bbl. of 40-gravity oil or better.

Jan 3, 1924

By Wilhelm Borchers

IN all special branches of the chemical and metallurgical industries, in which large electric furnaces became necessary for carrying out new processes or for the improvement of old ones, the development of the electric furnaces took place on much the same lines. The endeavors to make a success of a new idea brought forth many ingenious designs, which, however, almost without exception, furnished the proof that the safest and shortest way to practical success is the utmost simplicity of construction of the necessary apparatus. Our knowledge of the qualities of materials for building electric furnaces and of the constituents of the furnace-charges, especially their electric conductivity and their chemical affinity . towards one another at high temperatures, is still far from being complete, and from this fact we must conclude that the smaller the number of elements in the construction of a furnace, the fewer the number of unreliable factors both in the construction and in the operation of the electric furnace. This conclusion deserves special attention when an electric furnace for making steel has to be chosen from the constantly growing number of inventions, inasmuch as the chief con¬stituent of steel, the element iron, is a substance which in its liquid phase has so great an affinity fir those elements which are to be removed from the charge, that it is very difficult to comply with the following conditions: 1, to beat the refining-slag to the highest point of its chemical activity ; to prevent the iron from entering the slaw, or, where this cannot be done, to make it enter so that it finally appears in the end-product

Jan 1, 1910

By Walter W. Bradley

FOR a number of years up to the economic setback of the 1929-1931 period, the greatest proportional advances in the mineral industries in California were made among the substances in the nonmetallic groupings, particularly among the so-called industrial minerals and the salines. During the past year or two, although these have not regained the stride that marked their progress in the preceding decade, we can still record, -some new developments. California produces annually between 55 and' 60 different minerals on a commercial scale. Of these, some 40 to 45 comprise the "structural," "industrial," and "salines" materials as grouped in the reports of the State's Division of Mines. Barite-The barite deposits of California were described in Report

Jan 1, 1935

By CHARLES R. KETES

I. INTRODUCTORY. INDUSTRIALLY, the most important service that geological science can now render to mining in the Upper Mississippi leadand zinc-fields is to devise some practical scheme whereby the productive ore-belts may be located with greater certainty than they have been in the past. Up to the present time there has been little service of this kind, and in spite of all that has been worked out concerning the geologic features of the region, and notwithstanding all that has been written about these ore-deposits, this particular phase of the subject has been, for some reason or other, the very one that has received the least consideration. The neglect by geologists in this respect has been all but complete. The following notes are suggestive of some practical solutions to the problems presented. As here suggested, the geologic relationships of the ore-bodies were surmised some years ago, though somewhat vaguely perhaps. No opportunity for securing critical data appeared until recent investigations of a private character were undertaken, which bad in view the extensive prospecting of undeveloped areas. Fresh from rather wide experiences of a decade in the principal mining-regions of four continents, and after a lenstrum's residence among the chief mining-camps of the West and of Mexico, under conditions which measured geologic hypotheses in meters and rigidly tested their values in, terms of dollars, the recent return to the Joplin field and a new survey of its ore-conditions were fraught with somewhat unexpected results. Aspects of the ore-deposition were presented of which previously there was no hint. Ten years' absence from the district had tended greatly to clarify former hazy impressions. It enabled many features, then only dimly outlined, to be considered more nearly in their true perspective and in their real relationships to their surroundings. It permitted a great mass of details collected a dozen years before in connection with the work of the Missouri Geological Survey to be viewed under new light and to

Feb 1, 1909

By AIME AIME

The American Society of Mechanical Engineers.-The semi-annual meeting of the American Society of Mechanical Engineers will be held in Detroit, Mich., June 23-26. A session will be devoted to hoisting- and conveying-machinery; and many other subjects will be treated in professional papers, among which are . Clutches (with special reference to automobile clutches); The Thermal Properties of Superheated Steam; The Horse-Power, Friction-Losses, and Efficiency of Gas- and Oil-Engines; A Journal-Friction Measuring-Machine; A Simple Method of Checking Conical Pistons for Stress, and The By-Product Coke-Oven. The usual receptions will be held and excursions will be made to manufacturing plants, the slip-building yards, and various points of interest in and around Detroit. Among the excursions' planned is one to the University of Michigan at Ann Arbor. The Gas-Power Section of the Society will hold a session, and the Society for the Promotion of Engineering Education and the Society of Automobile Engineers will hold meet-

Jan 5, 1908

By Sidney Sigel, J. Gardner Brooks, Irvin R. Kramer

In 1942 Grossmannl proposed that the hardenability of a steel may be calculated from its chemical composition by considering the base hardenability associated with its carbon content and grain size and multiplying this base by factors for each element present. Since Grossmann's original publication, several investigaton2-6 have reexamined the multiplying factors for the individual elements and in general have found that the multiplying principle is valid. However, some of the individual factor curves developed have been in poor agreement. Inasmuch as the carbon-factor curve obtained by Grossmann was dependent Lipon the factors for all the other alloying elements present in the steel, it logically followed that his results were in need of correction in the light of the new data available for manganese, silicon, and the other alloying elements. Furthermore, in order to avoid the possibility that errors in the manganese, silicon, or other curves might influence the carbon-factor curve, investigation of iron-carbon alloys containing little or no manganese and silicon was necessary. That the effect of carbon on hardenability was greater than indicated by Grossmann is shown by a comparison of the diameters of completely hardened rounds of iron-carbon alloys with the ideal critical diameter calculated from Grossmann's factors. These alloys have a greater hardenability in a finite quench than predicted by Grossmann's factors for an infinite quench (Table I). Among the other important variables that were in need of further investigation were the effect of deoxidation practice on the grain-size correction curves and the effect of stable carbide-forming elements on the depth of hardening. Most Of the work by Kramer, Hafner, and Toleman4 was done on silicon-killed steels. It was not known whether Gross-mann's grain-size correction curves, determined on aluminum-killed steels, was valid when applied to steels with different deoxidation practice. More data on the

Jan 1, 1947

By J. Gardner Brooks, Sidney Sigel, Irvin R. Kramer

In 1942 Grossmannl proposed that the hardenability of a steel may be calculated from its chemical composition by considering the base hardenability associated with its carbon content and grain size and multiplying this base by factors for each element present. Since Grossmann's original publication, several investigaton2-6 have reexamined the multiplying factors for the individual elements and in general have found that the multiplying principle is valid. However, some of the individual factor curves developed have been in poor agreement. Inasmuch as the carbon-factor curve obtained by Grossmann was dependent Lipon the factors for all the other alloying elements present in the steel, it logically followed that his results were in need of correction in the light of the new data available for manganese, silicon, and the other alloying elements. Furthermore, in order to avoid the possibility that errors in the manganese, silicon, or other curves might influence the carbon-factor curve, investigation of iron-carbon alloys containing little or no manganese and silicon was necessary. That the effect of carbon on hardenability was greater than indicated by Grossmann is shown by a comparison of the diameters of completely hardened rounds of iron-carbon alloys with the ideal critical diameter calculated from Grossmann's factors. These alloys have a greater hardenability in a finite quench than predicted by Grossmann's factors for an infinite quench (Table I). Among the other important variables that were in need of further investigation were the effect of deoxidation practice on the grain-size correction curves and the effect of stable carbide-forming elements on the depth of hardening. Most Of the work by Kramer, Hafner, and Toleman4 was done on silicon-killed steels. It was not known whether Gross-mann's grain-size correction curves, determined on aluminum-killed steels, was valid when applied to steels with different deoxidation practice. More data on the

Jan 1, 1947

By McHenry Mosier

SUMMARY MERCURY is one of the strategic metals of which the supply has been raised from critical uncertainty to more than enough for essential demands. Work by the Bureau of Mines has contributed substantially to increased production of mercury in the United States and to consequent accumulation of Government stocks. Other contributing factors have been higher prices for mercury and Government loans to operators. The Bureau has undertaken 13 exploration projects, 10 of which have been completed. This work has comprised 28,376 ft. of several types of drilling and 23,798 lin. ft. of surface trenching, and has indicated commercial ore containing 34,800 flasks of mercury and marginal ore containing 26,200 flasks, a total of 61,000 flasks; equivalent, respectively, to domestic production for 8 months, 6 months, and 14 months. As Government stocks are now adequate, domestic production (currently at the rate of 50,000 flasks per year) is gradually being curtailed through Government control of labor and operating supplies.

Jan 1, 1944

By David R. Mitchell

Many of the difficult operating problems of the preparation of coal for market, of sampling coal shipments and in the utilization of coal are caused by segregation in the coal mass. Segregation may be of two kinds: i.e., "size segregation," in which particles of the same size in a mass of coal containing a range of sizes collect together and "density segregation," in which particles of the same density or specific gravity in a mass of particles of varying density gather together. Mechanics of Segregation Size segregation cannot take place with closely sized coal. As the size range increases, the tendency for size segregation increases. Similarly, cleaned coal with particles of approximately the same specific gravity have very little tendency for density segregation. The wider the density range of particles and the more particles of different density, the greater the chances are for density segregation. The following discussion deals primarily with coal of a wide size range as screenings and of wide density differences as uncleaned or run-of-mine screenings. If a mass of coal containing particles of a wide size range is subject to the stream action of a moving current of water, the small particles are transported farther than the coarser. Particles of different density in a stream of moving water are subject to the laws of classification, the heavier settling out first. The familiar sluice box of the gold miner and the action of rivers and other streams in classifying material according to

Jan 1, 1938

By David R. Mitchell

Many of the difficult operating problems of the preparation of coal for market, of sampling coal shipments and in the utilization of coal are caused by segregation in the coal mass. Segregation may be of two kinds: i.e., "size segregation," in which particles of the same size in a mass of coal containing a range of sizes collect together and "density segregation," in which particles of the same density or specific gravity in a mass of particles of varying density gather together. Mechanics of Segregation Size segregation cannot take place with closely sized coal. As the size range increases, the tendency for size segregation increases. Similarly, cleaned coal with particles of approximately the same specific gravity have very little tendency for density segregation. The wider the density range of particles and the more particles of different density, the greater the chances are for density segregation. The following discussion deals primarily with coal of a wide size range as screenings and of wide density differences as uncleaned or run-of-mine screenings. If a mass of coal containing particles of a wide size range is subject to the stream action of a moving current of water, the small particles are transported farther than the coarser. Particles of different density in a stream of moving water are subject to the laws of classification, the heavier settling out first. The familiar sluice box of the gold miner and the action of rivers and other streams in classifying material according to

Jan 1, 1938

By C. R. Van Hise

PART I.-GENERAL PRINCIPLES. [ ]

Jan 1, 1902

By William B. Millar

IN ONTARIO development of the gold mines is being rapidly pushed, while the intensity of the search for new mines has probably not been equaled at any time in the past. Even to outline the results of recent prospecting activity in Ontario would be a big undertaking, but it would be timely and, I hope, interesting to Institute members to review some generalities regarding the work of the prospector and then to discuss in some detail the growth of Ontario's most promising prospect area, the Swayze district, where serious mining operations are already under way. Probably no one writing about prospectors and pros¬pecting can resist hoping that his article will not reach the eyes of a certain portion of the uninformed general public and cause a further influx of "prospective prospectors" into any given area. It is somewhat mystifying how so many people reach the conclusion that prospecting is in

Jan 1, 1933

By F. F. Hintze

Petroleum production in the Rocky Mountain district during 1929 registered a small decline from that of the previous year, conforming with a gradual decrease in the amount of oil produced during the last few years. A number of causes may be mentioned to account for this result, chiefly overproduction in other regions and the apparent desire of operators in this territory to conform to the policy of proration and retrenchment adopted by special agreement among operators in many of the fields. Unit operation has been adopted to control oil production from the Sundance sands in the Salt Creek field, Wyoming, in line with the general effort to stabilize this phase of the industry and prevent excessive drilling of new wells. Another factor which greatly retarded development in the Rocky Mountain region was the executive order of President Hoover issued in March refusing the award of new prospecting permits to future applicants and threatening cancellation of all existing permits upon which actual drilling had not been done as specified by law. Only four states in the Rocky Mountain region are important oil producers. In 1929 Wyoming led with an average daily output of 52,600 bbl., followed by Montana, with 10,730 bbl.; Colorado, with 6500 bbl., and New Mexico with 2700 bbl. daily average. These states bear the following percentage relation to total production in the Rocky Mountain district: Wyoming, 72; Montana, 15; Colorado, 9; New Mexico, 4., (Table 1.) The total average daily production was approximately 73,000 bbl. and the total production of all four states during the year was 26,534,606 bbl. This is 2.7 per cent. of the total oil production of the United States during this period. Colorado Although Colorado is not a large producer of oil and natural gas, its oil operations are widely distributed over the state. The Florence field is one of the oldest oil fields in the country, and there are good reasons for believing that the state will retain its place for many years to

Jan 1, 1930

By C. E. Coleman, D. Hardie, A. E. Wraith

A photoguaplric method of measuring crack velocities has been applied to an examilzation of- the effect of hydrogen on the impact properties of zirconium. Results indicate that the crack in hydrided zirconit travels at about 26 m per sec whereas that in the uacuum -annealed is dependent on the speed of the hammer. RECENT work on the effect of hydrogen on zirconium reveals that when hydrided zirconium is tested in slow bending the crack producing failure can be unstable. The mode of fracture seems to involve the crack linking up broken hydrides by a ductile mechanism in the zirconium matrix.2" In order to compare the fracture of zirconium with that of other metals which show unstable crack propagation, notably steel, the rate of the crack spread must be known. As the methods previously used for measuring crack growth are either unsuitable for the present work5-'' or very costly," a simple technique has been devised in which the maximum crack velocity is derived from a series of crack position-time measurements. EXPERIMENTAL PROCEDURE The experimental arrangement is shown schematically in Fig. 1. The flash unit12 was triggered when the Charpy hammer touched the specimen and the 6-psec flash of the lamp delayed by a preselected time interval in the range 100psec to 10 msec. The camera was left on open shutter during the test. From direct measurement on a calibrated photographic print, Fig. 2, the position of the crack and the specimen was found at a known time after impact and, from a series of such tests using an appropriate range of flash-delay times, the crack velocity was estimated. Preliminary experiments were carried out on mild steel above and below its transition temperature. Vacuum-annealed zirconium, and zirconium hydrided to 100 ppm H were then tested at 20" and 150°C. For comparison, Zr-2.6 pct Nb alloy (in the 0 + transformed /3 condition), vacuum-annealed or hydrided to 100 ppm H, was tested at 20 C. The chemical analyses of these materials are given in Table I. For the high-temperature tests a fine chromel-alumel thermocouple was spot-welded onto each specimen close to the notch. The test piece was heated up to 200°C to allow enough time for preparing the apparatus while the specimen cooled to the testing tem- perature. It was assumed that the specimen cooled uniformly. The accuracy of the flash-delay time was better than i2 pct i 5 psec. The crack length was taken as its maximum distance traveled normal to the length axis of the specimen. Errors in these measurements arise as a result of the uncertainty in the location of the crack tip and the difficulty in defining the exact position of the crack starting point. The latter was due to the bonding and large amounts of deformation at the hammer contact point, especially in the vacuum-annealed specimens tested at 150°C. The maximum error in the measurements was however estimated to be about 5 pct. DISCUSSION OF RESULTS From the trial experiments using mild steel above its transition temperature it can be seen from the plot of crack length against time after impact, Fig. 3. that a reasonable estimate of the maximum crack velocity at the surface of the specimen may be made. Other features of note are the initiation time before the crack is first seen and the slowing down period

Jan 1, 1967

No. 264. A long established company operating steel and iron foundries, machine and forge shops, whose varied products are sold to the Government, railroads, mining and contracting industries, etc., has, openings in both manufacturing and selling departments for men with technical education and experience who are not subject to military call. State full particulars, kind of position preferred and salary expected. No. 265. Young man wanted, technical graduate preferred, to superintend construction f a small charcoal blast furnace. Man with blast-furnace experience preferred. For the right kind of man this is, not only a permanent job but an opportunity. In writing, give full particulars as to age, education, experience, salary wanted, etc. No. 26.6. An important copper mining and smelting company in Arizona needs two or three metallurgical chemists, at salaries. ranging front $150 to $175 per month. No. 267. A consulting mining engineer needs two superintendents for developing chrome deposits near Philadelphia; and another technical man for other engineering work in the same vicinity. No. 268. An iron-mining company, in New York State needs a mining engineer with knowledge of geology and acquaintance with construction work; A married man, with about 10 years' engineering experience preferred.

Jan 2, 1918

By C. E. Lesher

Two processes for agglomerating fine sized ores with low temperature coke are described. One process (Orcarb) agglomerates ores with limited amounts of carbon; the other (ore-carbon pellets) pelletizes fine sized ores, using low temperature cokes as the binder. Data are presented on the products obtained when taconite, magnetite, and hematite concentrates and several titanium oxide ores were used. REDUCTION of finely divided oxide ores has long been a major metallurgical problem. Various methods of agglomerating, notably briquetting, sintering, nodulizing, and pelletizing, have been developed and are in industrial use. Carbon is the usual reducing agent for oxide ores and attempts have been made to agglomerate fine ores with coking coal in order to get the metallurgical advantages of intimate contact and increased particle sizes suitable for subsequent smelting or reduction.1-3 For the past three years, research has been in progress on two processes for combining fine sized ores with reactive carbon in the form of low temperature coke. By one process, designated Orcarb, the ore is coated with coke carbon in amounts limited to that required for reduction and the result is an appreciable but limited increase in product particle size over that of the original ore. The second process, called ore-carbon pellets, produces pellets of fine ore bound by low temperature coke. The first objective was to develop a process for combining an oxide ore with only enough carbon for its reduction, the amount being that derived from the simple reduction equations. The carbon (that amount theoretically required for the reduction of the oxides in an agglomerate with ore), calculated to CO, ranges from 10 pct for zinc calcines Or phosphate Ores to 21 pct for rutile and 26 pct for silica. The carbon required for reducing iron oxides falls between the extremes—10 and 26 pet' When the carbon in the agglomerate is low temperature coke, reduction by hydrogen may be expected to take place, thus modifying the calculated carbon require- ment. On the other hand, in practice, an excess of reducing agent over that required in theory is normally used. Screen sizes of three iron ores on which the greater part of the experimental work was performed are given in Table I. Most of the testing was done with Freeport (Renton Mine) coal, but Pittsburgh and Elkhorn bed coals were also used. The first objective was to agglomerate the ore with sufficient low temperature coke to provide carbon for reduction, i.e., in the range from 10 to 25 pct. This was accomplished by preheating the ore, mixing with pulverized coking coal, and completing the coal carbonization in a plain steel rotating retort. The pilot retort, as it is now constituted, is shown by diagram in Fig. 1 and photograph, Fig. 2. Ore is fed into the upper kiln by a screw; it is heated to 900° to 1100°F by direct flame and it is then discharged into an insulating hopper from which it goes

Jan 1, 1956

By C. E. Lesher

Two processes for agglomerating fine sized ores with low temperature coke are described. One process (Orcarb) agglomerates ores with limited amounts of carbon; the other (ore-carbon pellets) pelletizes fine sized ores, using low temperature cokes as the binder. Data are presented on the products obtained when taconite, magnetite, and hematite concentrates and several titanium oxide ores were used. REDUCTION of finely divided oxide ores has long been a major metallurgical problem. Various methods of agglomerating, notably briquetting, sintering, nodulizing, and pelletizing, have been developed and are in industrial use. Carbon is the usual reducing agent for oxide ores and attempts have been made to agglomerate fine ores with coking coal in order to get the metallurgical advantages of intimate contact and increased particle sizes suitable for subsequent smelting or reduction.1-3 For the past three years, research has been in progress on two processes for combining fine sized ores with reactive carbon in the form of low temperature coke. By one process, designated Orcarb, the ore is coated with coke carbon in amounts limited to that required for reduction and the result is an appreciable but limited increase in product particle size over that of the original ore. The second process, called ore-carbon pellets, produces pellets of fine ore bound by low temperature coke. The first objective was to develop a process for combining an oxide ore with only enough carbon for its reduction, the amount being that derived from the simple reduction equations. The carbon (that amount theoretically required for the reduction of the oxides in an agglomerate with ore), calculated to CO, ranges from 10 pct for zinc calcines or phosphate ores to 21 pct for rutile and 26 pct for silica. The carbon required for reducing iron oxides falls between the extremes—10 and 26 pct. When the carbon in the agglomerate is low temperature coke, reduction by hydrogen may be expected to take place, thus modifying the calculated carbon require- ment. On the other hand, in practice, an excess of reducing agent over that required in theory is normally used. Screen sizes of three iron ores on which the greater part of the experimental work was performed are given in Table I. Most of the testing was done with Freeport (Renton Mine) coal, but Pittsburgh and Elkhorn bed coals were also used. The first objective was to agglomerate the ore with sufficient low temperature coke to provide carbon for reduction, i.e., in the range from 10 to 25 pct. This was accomplished by preheating the ore, mixing with pulverized coking coal, and completing the coal carbonization in a plain steel rotating retort. The pilot retort, as it is now constituted, is shown by diagram in Fig. 1 and photograph, Fig. 2. Ore is fed into the upper kiln by a screw; it is heated to 900" to 1100°F by direct flame and it is then discharged into an insulating hopper from which it goes

Jan 1, 1956

By AIME AIME

ARRANGEMENTS for the Annual Meeting of the Institute were well advanced at the end of December as the following program will show. Heretofore this has been printed separately, but its inclusion in the Annual Review issue of MINING AND METALLURGY will, it is hoped, make it more easily available for reference until the final program is issued, which will be distributed at the meeting. Added details will appear in the February issue. Last year's meeting drew a record crowd of over 2300 which should be exceeded this year. Early reservations at the headquarters hotel, the Waldorf-Astoria, or any other convenient hotel are desirable, as are prompt reservations for the Annual Dinner. No one need worry about too much speechmaking this year, for the honors will be awarded and accepted in a few words instead of a few paragraphs, in seconds rather than minutes.

Jan 1, 1939

By J. H. Swisher

ALUMINUM deoxidation equilibrium in liquid iron has been the subject of many investigations. Sawamura and Sano1 have written a critical survey of the literature on this subject and consider the data of Gokcen and chipman2 and of d'Entremont, Guernsey, and Chip-man3 to be most reliable. The present study was undertaken to check these data and extend the range of the experiments to lower temperature. The experiments consisted of equilibrating 5-lb Fe-Al melts with recrystallized alumina crucibles at 1580o + 10oC, and analyzing the solidified ingots for aluminum and oxygen. The electrolytic iron stock contained 100 ppm C, 20 ppm 0, and trace amounts of other elements.* The stock was induction-melted in an atmosphere of 5 pet H2 in argon to attain equilibrium between the melt and the crucible. The temperature was measured with a W, 3 pet Re/W, 26 pet Re thermocouple, and the equilibration time was about 4 hr. The results are plotted in Fig. 1 along with curves obtained from the data of d'Entremont et al.,3 for temperatures of 1740o and 1910°C. The present results can be seen to be consistent with the previously reported high-temperature data. The reaction under study is AI2O3 (s) = 2 Al (l) +3O [1] for which the deoxidation product is Kl =[% Al]2 [%O]3 [2] (Under the experimental conditions, Fe A12O4 is not expected to form.4) In Fig. 2, values of log Ki evaluated at 0.01 pet A1 are plotted against 1/T, using data obtained from all the investigations under consideration. The deoxidation product is plotted in Fig. 2 in lieu of the equilibrium constant (normally obtained by evaluating Ki as pet A1 — 0) because of the uncertainty involved in extrapolating to zero concentration. The difficulty in extrapolating the present data for 1580°C is apparent curvature in the log Ki vs pet Al plot. This curvature is perhaps due to the uncertainty in oxygen analysis for concentrations of 5 ppm and below. For the data of Gokeen and chipman,' the extrapolation is also difficult because of experimental scatter. The line drawn through the points plotted in Fig. 2 may be represented by the equation:

Jan 1, 1968