Search Documents

By Benjamin LeRoy Miller, Joseph T. Singewald

The ore deposits at Firmeza have been worked continuously since 1884; those at Daiquiri since 1895. It is surprising, therefore, that they have not been the object of careful geologic study until quite recently. It is true that considerable attention was paid to these deposits by A. C. Spencer, in 1901, in the report on the geology of Cuba, by Hayes, Vaughan, and Spencer,l but this was only incidental to a hasty general reconnaissance of the Island. As far as the Daiquiri deposits are concerned, this gap has been filled by the appearance of the two papers by Prof. James F. Kemp2 and by Prof. Waldemar Lind-gren and Clyde P. Ross respectively.3 Max Roesler, first holder of the Emmons fellowship, made a detailed study of the Firmeza deposits during the past summer, so that soon we can expect a full description of them. Consequently, there is no, demand at the present time for a description of the deposits from other sources, and that is not the purpose of this paper. The papers on Daiquiri, however, leave unsolved problems in regard to the age of the rocks of the district and certain phases of the genesis of the deposits, and it is with the hope of aiding in the solution of some of these problems that this contribution is offered. Kemp, and Lindgren and Ross agree in assigning these deposits to the type of contact metamorphic deposits. They disagree in part on the question of their localization. The deposits occur in an extensive area of diorite which includes numerous blocks of limestone, and sometimes are found in the contact zones between these blocks and the diorite, and in other instances are completely enveloped in the diorite, showing no apparent connection with the limestone. The latter class includes

Jan 1, 1916

By W. Mostowitsch, H. O. Hofman

The mineral gypsum, CaSO4 + 2 H2O, has been used for many years as a sulphurizing and basic flux in several smelting-operations. Thus, in smelting oxide nickel-ore in the blastfurnace, it is commonly added to the charge to furnish the sulphur necessary for collecting the metal in a matte, and a base for slagging the siliceous gangue. In the concentration of lead-copper matte in the reverberatory furnace it has been used for years at Preiberg, Saxony,1 for a similar purpose, and for producing at the same time a copper-matte with less than 0.15 per cent. of Fe. The latest use gypsum has been put to is in the blast-roasting process of Carmichael-Bradford.2 The term " blast-roasting," given by A. S. Dwight3 to the Dwight-Lloyd method of roasting and agglomerating: is a happy generic term which covers the ground better than the " lime-roasting " of Ingalls5 or the " pot-roasting " of Austin: in that it leaves the operation independent of the character of the flux and the form of apparatus, and retains the characteristic feature of this class of processes—namely, that of using forced draft. In the Carmichael-Bradford process the dehydrated material, mixed with galena-concentrate, acts as a diluent and a flux. The process differs in this from the Huntington-Heberlein7 and the Savelsberg8 processes, in both of which limestone is used.

Jan 1, 1909

By J. V. N. Dorr

THE Elko Prince mine is in the Gold Circle district, Nevada, about ½ miles(2. 4km.) from the town of Midas, 55 miles (88.5 km.) west of Battle Mountain and 50 miles (8.5 km.) northeast of Golconda. The district is described by Emmons in U. S. Geological Survey Bulletin 408, 1910. Little was said of the Elko Prince, however, as only a small amount of work had then been completed. The geological structure of the region is simple. The oldest exposed formation, covering most of the area, is rhyolite, which has a light-colored devitrified groundmass enclosing phenocrysts of feldspar, quartz, and magnetite. It has been somewhat weathered and partially stained with oxide of iron. There are several outcrops of andesite which cut the rhyolite. The ore deposits occur as fissure veins, replacement veins, and sheeted zones. The Elko Prince vein is of the first type. The rhyolite near the veins is silicified, stained with oxide of iron, and, at some localities, replaced by ore, or is so permeated by metals as to be classed as ore. The vein filling consists chiefly of quartz, and the values consist entirely of gold and silver. Part of the gold exists as the native metal and part is associated with pyrite. Silver occurs free and its argentite in handed streaks through the vein. These veins trend in a northwesterly direction, with dips varying from 65° to vertical.

Jan 8, 1918

By G. T. Brookman, P. B. Katz, D. K. Martin, J. A. Ripp

With the increasing emphasis on using American fossil fuel resources, more facilities are evaluating coal conversion. In addition, those facilities now using coal are considering better coal storage procedures to minimize negative environmental impacts. For example, rainfall and snowmelt on a coal pile produce a drainage stream high in acidity, suspended and dissolved solids, iron, sulfate, and trace metals. Usually, the characteristics of the runoff stream do not allow for direct discharge to streams and lakes. Detention in a holding pond and pH adjustment are traditional methods of treating coal pile runoff before discharge. Historically, this treatment methodology has been performed inefficiently without a good data base to assess how much runoff treatment is needed to meet effluent guidelines. A model was developed to simulate runoff volume and pollutant loadings under different rainfall conditions. This model will be used to size and design treatment works for coal pile drainage. The Ohio State University (OSU) version of the Stanford watershed model was selected as the base model. The OSU model simulates overland runoff in a rural area. It includes factors such as soil moisture, depression storage, infiltration, and flow to a ground water reservoir. TRC also used the OSU co-model for acid production in coal refuse piles as the basis for qualitative modeling work. A number of modifications were made to the OSU models to make them applicable to coal storage piles. The physical/chemical reactions included in the models are described in more detail in the project report. In addition, the models were altered so that a standard NOAA magnetic tape of historical meteorological data can be used as data input instead of requiring the user to create a data file. The resulting TRC coal pile drainage model is actually two models - a quantitative model, termed H2OTRC, and a qualitative model, TRCCOAL.

Jan 1, 1984

JOHN W. BELL,* Montreal, Canada (written discussion?).-Mr. Hardinge's "hammer" theory is, I believe, a sound one; but, unfortunately, at the very place in the Hardinge mill where there should ,be a large number of small "hammers," there is room for only a very few. The results in Mr. Van Winkle's paper seem to indicate that the practical development of the theory can best be accomplished by stage crushing, using cylindrical mills. It would be very interesting to know if any stage-crushing experiments have been made at Inspiration with Marry mills. Mr: H. Kenyon Burch says:1 "The Marcy mill also proved to be a most excellent intermediate crusher, handling as high as 800 tons of 3-in. feed and under, to pass an 8-mesh screen without return of oversize." It seems reasonable to suppose that a large part of this output would be a finished- product; consequently, if it were possible to send the oversize from two of the primary mills to one secondary mill using balls of small diameter, a considerable saving in power might be realized, provided it were not necessary greatly to enlarge the secondary mill. It does not seem reasonable to suppose that fine material can he crushed as quickly or as efficiently in a primary ball-mill as in a re-crushing mill especially designed for that work. Experimental data, of course, are essential in deciding questions of this kind.

Jan 4, 1918

By R. S. Rickard, B. Ball

INTRODUCTION Pyrite responds to the flotation reagents normally used in sulfide flotation systems. For example, it is well known that xanthates are good collectors for pyrite, and cyanide, sulfide and hydroxides are depressants. The particular response of pyrite to these and other reagents is used to advantage in several areas. By careful control of depressant and collector additions, pyrite can be separated from other sulfides by flotation. This is of obvious importance in base metal sulfide flotation where pyrite depression is often required to produce a marketable grade concentrate. Flotation of pyrite has long been employed in gold ore processing, and it appears that pyrite flotation from coal may become important in the future (1). It is probably a direct consequence of this relative ease of separation, plus the low economic value of pyrite, that detailed investigations on the chemistry of its flotation and depression are relatively limited. Gaudin (2, 3, 4) and others, in the mid-1950's published data on the flotation of pyrite, but detailed or consistent mechanisms were lacking. In 1955, Sutherland and Wark's book (5) was published, in which a large , amount of empirical information, especially earlier work of Wark and Cox, was presented. More recently, in 1968, significant contributions have been made by Fuerstenau et al. (6, 7) and Majima and Takeda (8). These papers introduced electrochemical explanations of the effects observed with xanthate and cyanide ions.

Jan 1, 1976

By W. A. Krivsky, R. Schuhmann

PREVIOUS papers in this series on the thermo-•t dynamics of copper-smelting systems have presented a survey of the field and an outline of the overall program: thermodynamic studies of iron silicate slags, 2-4 and thermodynamic data for simple mattes in the Cu-S system.5 Since this program was started, an excellent summary of the whole field of physical chemistry of copper smelting was published by Ruddle. The present paper represents a further step in the study of liquid mattes, and deals specifically with the Cu-Fe-S system. The importance of the Cu-Fe-S system to copper smelting is evident from the fact that these three elements constitute 90 to 99 pct of the totalweight of most commercial mattes. In copper mattes oxygen is an important fourth component. Oxygen was rigorously excluded in the work to be described in this paper. However, the behavior of oxygen in mattes was the subject of a companion research by Lander' which will be described in a subsequent publication. Several studies have been made previously of phase equilibria in the Cu-Fe-S system,8-10 so that the ternary phase diagram is fairly well established in most regions. Also, separate studies have been made of the phase diagrams and thermodynamic properties for the three binary systems: CU-S,5,12-15 CU-Fe,11,14,17 and FeS.11,14,16,17 However, no work has

Jan 1, 1958

By J. C. Jaeger

It frequently happens in engineering and mining situations that "bad" rock is encountered which consists of rock broken up into blocks a few inches or less in diameter by a network of intersecting joints. These joints may be open, or filled by veins of solid material, or covered with weathering products. Essentially, the material consists of an aggregate of small interlocking blocks. In such circumstances, major joint systems usually can be recognised and studied, but there are probably many minor planes of weakness which may cause the blocks to be of almost random shape. It is extremely difficult to determine the properties of such materials. Measurements on small specimens give the properties of the constituent rock and not the jointed material. At the other extreme, the material has occasionally been tested as loose aggregate in large triaxial pots, but this gives no information about the strengthening effect of the interlocking of the small blocks of rock. One simple approach, which has been much used, is to regard the rock as randomly jointed on a small scale and to apply soil mechanics theory to it. In this case, the normal and shear stress at failure are related by [ ] where the coefficient of friction p and the cohesion c will be those of the individual joints. Even in very bad rock, it is usually possible to drill core in small diameters of 1 to 2 in. which will contain a number of joints, both open and filled. The testing of such joints provides one simple method of determining some of the properties of closely jointed rock and will be discussed in detail in the next section. Calculations based on Eg. 1 with values of c and µ obtained in this way may be expected to give a lower limit for the strength of structures in rock.

Jan 1, 1970

By P. St. J. Perrott

AN explosive of balanced composition might be expected, on first consideration, to develop maximum propulsive strength.1 This hypoth¬esis is based on the assumption that an excess of oxygen would act simply as an inert material with a consequent lowering of flame temperature, also that a deficiency of oxygen would tend toward incomplete reaction of the combustible materials, thereby preventing development of the full strength of the explosion. Actually, however, it can be shown both theoretically and experimentally that an explosive having a certain oxygen deficiency will develop the greatest propulsive strength. This fact has been known to the technical staffs of explosives manufacturers for many years, but it has not always been recognized by explosives engineers in the field. For example: There is a fairly widespread opinion that the wrapper of a stick of dynamite is merely a container for the explosive and that it plays little if any part in the development of propulsive strength. Actually, it can be shown that the wrapper enters fairly completely into the explosive reaction. In the case of most gelatin dynamites, a given weight of explosive from which the wrappers have been removed will develop considerably less strength than the same weight of explosive plus wrapper, in spite of the fact that the latter combination contains less weight of the actual explosive ingredient. It can also be shown that a cartridge 7/8 in. in diameter may be stronger, weight for

Jan 1, 1928

By D. C. Buffum, L. D. Jaffee

THAT the path of brittle fracture in temper brit-tle steel differs from the path in unembrittled steel has been known for some years1 More recently it has been recognized that alloy steels, at least, display two forms of temper brittleness1 which may be called upper-nose and lower-nose. No experimental evidence has been published, however, to show that the two forms can be distinguished by simple inspection of brittle fractures. Fig. 1 is a photograph of the fractured surface of a V-notch Charpy specimen of SAE 3140 steel in which temper brittleness developed during heat treatment in the lower-nose temperature range.' The steel was austenitized 1 hr at 900°, tempered 1 hr at 675", and embrittled 48 hr at 500°C, with water-quenching after each treatment. The temperature of transition from ductile to brittle fracture in the Charpy test was + 2°C (criterion of transition: 100 pet fibrous). The hardness was Re 25; the austenite grain size ASTM 8. This specimen was broken at 150°C in the standard Charpy test and absorbed 6.6 ft-lb; its fracture was rated 0 pet fibrous. Fig. 1 shows that the fracture appeared rough to the naked eye. At the side of the specimen away from the notch (the side which was in compression during the test), the surface as a whole deviated sharply from a plane. At higher magnification, X60, the surface gave

Jan 1, 1959

By Stewart L. Toleman, Robert H. Hafner, Irvin R. Kramer

In his paper on the calculation of harden-ability from chemical composition, Gross-mannl discussed the effect of most of the alloying elements used commercially. The purpose of the work reported in this paper was to determine the effect of several other elements on hardenability—arsenic, antimony, beryllium, cobalt, columbium, germanium, tellurium, tin, aluminum, and titanium—and to extend the range of composition of some of the elements studied by Grossmann—chromium, copper, manganese, molybdenum, nickel and silicon. Experimental Methods To determine the effect on hardenability, suitable amounts of the alloying elements were added to experimental heats and the hardenability determined by means of end-quenched hardenability specimens. The steels were melted in a high-frequency induction furnace with a basic crucible, using ingot iron as a base for some of the heats and S.A.E. 1015 for others. The alloying additions were made as ferroalloys or as metal, depending on the alloying element. Hundred-pound heats were split and poured into square, tapered, cast-iron ingot molds with hot tops. Aluminum was not used as a deoxidizer in most of the heats. The ingots were homogenized for 12 hr. at 1260°C. (23oo°F.) and forged into rounds of I½ -in. diameter. These bars were normalized twice from 840°C. (I55o°F.) before machining into Jominy bars. mined in most Cases by samples taken from the end-quench hardenability spedmen. AS suggested by Crafts,2 acid-soluble (8

Jan 1, 1944

By Robert H. Hafner, Irvin R. Kramer, Stewart L. Toleman

In his paper on the calculation of harden-ability from chemical composition, Gross-mannl discussed the effect of most of the alloying elements used commercially. The purpose of the work reported in this paper was to determine the effect of several other elements on hardenability—arsenic, antimony, beryllium, cobalt, columbium, germanium, tellurium, tin, aluminum, and titanium—and to extend the range of composition of some of the elements studied by Grossmann—chromium, copper, manganese, molybdenum, nickel and silicon. Experimental Methods To determine the effect on hardenability, suitable amounts of the alloying elements were added to experimental heats and the hardenability determined by means of end-quenched hardenability specimens. The steels were melted in a high-frequency induction furnace with a basic crucible, using ingot iron as a base for some of the heats and S.A.E. 1015 for others. The alloying additions were made as ferroalloys or as metal, depending on the alloying element. Hundred-pound heats were split and poured into square, tapered, cast-iron ingot molds with hot tops. Aluminum was not used as a deoxidizer in most of the heats. The ingots were homogenized for 12 hr. at 1260°C. (23oo°F.) and forged into rounds of I½ -in. diameter. These bars were normalized twice from 840°C. (I55o°F.) before machining into Jominy bars. mined in most Cases by samples taken from the end-quench hardenability spedmen. AS suggested by Crafts,2 acid-soluble (8

Jan 1, 1944

By G. Kimmel, A. Bar-Or

The rate of nucleation and the rates of both lateral and longitudinal growth of martensite plates (needles) in b—a transformation were determined as a function of temperature, in various U-Cr alloys. The activation energies for nucleation and both modes of growth were calculated from the experimental results. The identical values for lateral and longitudinal growth clearly indicate that both are controlled by the WHITE,' Mott and Hines and Holdeen3 showed that characteristic martensite plates (needles) appear in U-Cr alloys during the transformation 8 —0 in the lower temperature range (up to -300°C). Holden pointed out that the rate of growth of the individual martensite plates under isothermal conditions is same mechanism. Lomer's suggestion that single-atom movement over subatomic distances controls the specially slow growth of martensite receives further support from the present kinetic study, which shows that the growth rate can be described quantitatively by known formulas for the growth rate in processes involving single-atom movement. very slow, and suggested that this is due to the slowness of the relief of the stresses engendered during the growth of the individual plates. lomer4 attributed the slow growth to the complexity of the atomic adjustment at the growing interface b/(u. In the present work the rates of nucleation and growth of martensite plates in U-Cr alloys were studied by direct observation, in order to elucidate the mechanism which controls their characteristic slow growth.

Jan 1, 1968

By W. C. Bleimeister

The rock salt deposits of the U. S. occur in five major basins: 1) the Eastern Basin (which includes parts of Michigan, Ohio, Pennsylvania, New York, West Virginia, and Ontario) ; 2) the Gulf Coast Basin (the northern rim of which includes parts of Alabama, Mississippi, Louisiana, Arkansas, and Texas, while its southern rim includes parts of the Isthmus of Tehuampetic and the Yucatan Peninsula of Mexico), 3) the Southwestern Salt Basin (which includes parts of Kansas, Colorado, Oklahoma, Texas, and New Mexico), 4) the Williston Basin (which under- lies parts of North Dakota, South Dakota, Montana, and Saskatchewan), and 5) the Green River Basin (whose periphery underlies parts of Wyoming, Utah, and Colorado). The greatest salt production in the U. S. is obtained from the Eastern Basin,-particularly from the areas of Michigan, Ohio, and Ontario. In 1958 the salt production from these areas amounted to 8.7 million tons or 35.3 pct of the combined total produced in the U. S. and Canada. Of the 24,286,000 tons produced by Canada and the United States, 6,443,000 tons were in the form of rock salt from dry mining, while 17,843,000 tons were produced from wells in the form of artificial brine, the result of dissolving the rock salt formation by fresh water. Three important factors lie behind the location and development of this salt mining area:

Jan 5, 1961

By Douglass D. Donald

The New Jersey Zinc Co. successfully holed through a 2 ½ -mile haulage tunnel connecting its new Ivanhoe shaft with the Van Mater Shaft at Austinville, Va. This 8 x 10-ft cross-section tunnel was driven from both ends and met at a point approximately a mile from one shaft and a mile and a half from the other. In order to make certain that the faces met accurately on line and grade it was necessary to refine conventional mine surveying practices. The mines that are serviced by these shafts are located in the southwestern part of Virginia approximately midway between Roanoke and Bristol at the west foot of the Blue Ridge. While the presence of lead-zinc sulfide ores has been known in this area for more than 200 years, it was in 1902 that New Jersey Zinc Co. acquired the Austinville property, which has been in constant operation since that time. About ten years ago the company undertook an extensive diamond drilling program at Ivanhoe several miles to the southwest. This program was successful in locating sufficient ore to justify the purchase of a number of the properties in this area and plans were made to develop the new orebodies. The discovery of this ore as well as substantial additional reserves at Austinville resulted in a decision by the company to increase its daily milling capacity to 3000 tons, of which about 1000 tons would be supplied from the Ivanhoe mine.

Jan 9, 1958

By H. C. Doepken

Wrought Hadfield steel was tested in axial tension at from 100° to —196°C, to determine flow and fracture stresses as well as conventional properties. Ductility and related properties, such as fracture stress, decreased continuously with temperature. Peculiarities during straining indicated possible martensite formation or mechanical twinning. THE mechanical properties of austenitic manganese steel have been summarized by Avery and Day,' including the low temperature tensile properties reported by Hadfield and coworkers2,3 " These data show the material to be brittle at both liquid oxygen and liquid 'hydrogen temperatures. Avery and Day report considerable ductility measured by Charpy V-notch tests at —100°F (-73°C). Since the review of low temperature properties by Seigle and Brick' show face-centered cubic structures, other than Hadfield steel, to be ductile at low temperatures, it was felt that a study of tensile properties, between the reported intervals, should show whether the decrease in ductility was a gradual decline or an abrupt drop beginning at some intermediate point. X-ray examinations were made in an attempt to shed further light on the mechanism of work hardening of austenitic manganese steel. The presence of a magnetic constituent has been found by Hall" and by Avery, Homerberg, and Cook." Krivobok7 demonstrated the growth of magnetic constituents on reheating cold-worked specimens. Hadfield2,3 ' and Goss found no a iron. The steel from which the specimens were taken was commercial hot rolled % in. diam rod of electric furnace steel. The ladle analysis was as follows: C, 1.40 pct: Mn, 12.11; P, 0.060; S, 0.012; Si, 0.120; Ni, 0.080; and Cr, 0.090. Fifteen specimens, 6 in. long, were cut from a

Jan 1, 1953

By L. Dudley Stamp

On April 1, 1937, Burma, formerly a province of the Indian Empire, was formally separated from India and became an independent unit of the British Commonwealth. In future, therefore, statistics of oil production in India will exclude Burma and those for Burma will be separately published. The last few years have been marked by noteworthy, almost spectacular, developments in the oil fields of Burma and the Punjab. In Burma these developments have been the proving of unexpected extensions of the existing fields and not the discovery of entirely new fields. The new extensions are important not only to Burma in giving a new lease of life to her fields, but also in general as stressing some interesting geological aspects hitherto neglected. Production The production of crude oil in Imperial gallons has been: Burma India 1934 254,760,070 67,265,210 1935 251,338,974 . 71,323,362 1936 265,656,393 69,241,504 1937 (eat.) 270,000,000 70,000,000 The Chindwin Fields.—The small field at Indaw continues to produce steadily about 3,000,000 gal. a year, but no really deep test has yet been successfully carried out and resources are thus not yet fully known. The remote location of the field, inaccessible except by river and situated 22 miles from the Chindwin 270 miles up that river from its junction with the Irrawaddy, has not encouraged active exploitation. The field is worked exclusively by the Indo-Burma Petroleum Co., Ltd. (I.B.P.), using now an automatic gas-lift system, and the company formerly refined the oil at a small refinery at Pantha on the river near by. The chief market was for kerosene in upper Burma. The refinery has now been closed. The oil is stored on the field during the low-water season

Jan 1, 1938

By G. St. J. Perrott

AN explosive of, balanced composition might be expected, on first consideration, to develop maximum propulsive strength.1 This hypothesis is based on the assumption that an excess of oxygen would act simply as an inert material with a consequent lowering of flame temperature, also that a deficiency of oxygen would tend toward incomplete reaction of the combustible materials, thereby preventing development of the full strength of the explosion. Actually, however, it can be shown both theoretically and experimentally that an explosive having a certain oxygen deficiency will develop the greatest propulsive strength. This fact has been known to the technical staffs of explosives manufacturers for many years, but it has not always been recognized by explosives engineers in the field. For example: There: is a fairly widespread opinion that the wrapper of a stick of dynamite is merely a container for the explosive and that it plays little if any part in the development of propulsive strength. Actually, it can be shown that the wrapper enters fairly completely into, the explosive reaction. In the case of most gelatin dynamites, a given weight of explosive from which the wrappers have been removed will develop considerably less strength than the same weight of explosive plus wrapper, in spite of the fact that the latter combination contains less weight of the actual explosive ingredient. It can also be shown that a cartridge 7/8 in. in diameter may be stronger, weight for

Jan 1, 1928

By V. W. Bihlman, H. M. Williams

In the construction of internal-combustion engines, the cooling of the combustion chamber and pistons is of great importance. In certain types of engines, the adequate dissipation of heat from the bearings is also important. One of the important factors affecting heat dissipation from these parts is the thermal conductivity of the material used. A search of the technical literature shows that, in general, the thermal conductivity of the elements1 has been determined accurately by a number of investigators. The data on the industrial alloys are, however, meager. For an investigation of these alloys, the thermal-conductivity apparatus described in this paper has been constructed at the laboratory of the General Motors Research Corporation. It is desirable that an apparatus for work of this nature should give fair accuracy of results with a convenient form of specimen and a reasonable rapidity of operation. A summary of the different methods of determining thermal conductivity is given in the first volume of the " Dictionary of Applied Physics," pages 429 and 470. The method selected as best satisfying these conditions is that of Griffith, on page 443, with the addition of a guard ring similar in principle to the method used by Berget (P. 445). Briefly, the method consists of heating one end of a bar T, Fig. 1, electrically, measuring the amount of heat transmitted through the bar by a flow calorimeter J fastened to the other end, and measuring the temperature gradient along the bar. To compensate for the loss of heat laterally, the test bar is surrounded by a guard ring E, in the form of a steel pipe, which is electrically heated at the bottom and cooled at the top with a

Jan 1, 1923

By V. W. Bihlman, H. M. Williams

In the construction of internal-combustion engines, the cooling of the combustion chamber and pistons is of great importance. In certain types of engines, the adequate dissipation of heat from the bearings is also important. One of the important factors affecting heat dissipation from these parts is the thermal conductivity of the material used. A search of the technical literature shows that, in general, the thermal conductivity of the elements1 has been determined accurately by a number of investigators. The data on the industrial alloys are, however, meager. For an investigation of these alloys, the thermal-conductivity apparatus described in this paper has been constructed at the laboratory of the General Motors Research Corporation. It is desirable that an apparatus for work of this nature should give fair accuracy of results with a convenient form of specimen and a reasonable rapidity of operation. A summary of the different methods of determining thermal conductivity is given in the first volume of the " Dictionary of Applied Physics," pages 429 and 470. The method selected as best satisfying these conditions is that of Griffith, on page 443, with the addition of a guard ring similar in principle to the method used by Berget (P. 445). Briefly, the method consists of heating one end of a bar T, Fig. 1, electrically, measuring the amount of heat transmitted through the bar by a flow calorimeter J fastened to the other end, and measuring the temperature gradient along the bar. To compensate for the loss of heat laterally, the test bar is surrounded by a guard ring E, in the form of a steel pipe, which is electrically heated at the bottom and cooled at the top with a

Jan 1, 1923