that it eliminates plastic deformation, that is, permanent set. Other references to important investigations are: HULL, W. A. "Fire Tests of Concrete ing concrete by this method. The surface-area method assumes that the physical properties of concrete are primarily dependent upon the relation of the volume of cementing material to the surface area of the aggregate and recognizes the importance of water as an essential ingredient of concrete. A. N. Talbot (ibid., xix, Pt. II) has developed a modification Gypsum.-Definite progress in of the surface area method involving standards for gypsum may now be rea function called "surface modulus" corded. "Specifications for Gypsum which is proportional to surface area. and Calcined Gypsum" and "Methods R. B. Young (ibid.) has endeavored for Tests of Gypsum and Gypsum to establish a relation between the two Products" have been issued tentatively methods. It seems evident, however, by the American Society for Testing that there is no definite relation be- Materials (Proceedings, xix, Pt. I). tween the fineness modulus and the A report on gypsum plasters specifysurface area of graded aggregate, and ing properties of materials and direcit remains to be established which tions for erection and workmanship, will better serve as a basis for pro- has also been published as a basis for portioning concrete from given aggre- future specifications, as well as a regates to produce concrete of specified port of recommendations relative to strength. Other references on this the use of reinforced gypsum in strucsubject are Engineering News-Record, tural design, similar to the report on June 19; and R. W. Crum, "Propor-reinforced concrete published in 1917. tioning of Pit-Run Gravel for Concrete" (Proc. Am. Soc. Test. Mats., xix, Pt. II). (See also Civil Engineering, supra.) Reference should also be made to Non-Ferrous Metals and Alloys.The standardization of non-ferrous metals and alloys noted in the last report (A. Y. B., 1918, p. 610) continues to be a significant development. Specifications for lead, brass ingot metal, bronze bearing metal, solder metal and tinned annealed copper wire, and methods for assay of copper and chemical analysis of pig lead have been issued tentatively by the American Society for Testing Materials (Proceedings, xix, Pt. I). Valuable data on the elastic properties of concrete, based upon several thousand compression tests, are contained in a paper by Stanton Walker (ibid.). Defining modulus of elasticity as the ratio of an increment of stress to a corresponding increment of deformation, Walker proposes two fundamental laws governing elastic properties of concrete: the first expresses the relation between stress and deformation as 8 Kd" where s= unit stress below the "yield point," dunit deformation, Ka con- Tests of light aluminium casting stant depending upon strength, and alloys are described by P. D. Merica na variable exponent; and the and C. P. Karr (ibid., Pt. II). Copsecond expresses the relation between per, zinc and magnesium are used to ultimate compressive strength and impart hardness and strength to modulus of elasticity as ECS", aluminium castings. Alloys containwhere E modulus of elasticity, Sing eight per cent. of copper are hard compressive strength, Ca constant and readily machined but are brittle, depending on conditions of the test, having an elongation in two in. of After testand ma variable exponent. In a only about two per cent. discussion of this paper S. C. Hollister ing several compositions, the authors proposes a different conception of conclude that an alloy containing only modulus of elasticity, as the ratio of from two to three per cent. of copper an increment of stress to the corre- with one or two per cent. of nickel or which has a tensile sponding increment of elastic deforma- manganese, tion, which he believes to be a better strength of 20,000-25,000 lb. per measure of elasticity of a material in square inch and an elongation of five = per cent., should have commercial pos- the steel the appearance of which sugsibilities. The addition of magnesium gests crystals "that were squeezed toto alloys containing copper reduces gether while plastic and yet failed to ductility but increases tensile strength adhere firmly to one another." They and hardness. Annealing increases appear to originate in the ingot, are the tensile strength and hardness but often associated with slag films, and generally decreases the ductility. The occur in both transverse and longiresistance of aluminium alloys to cor- tudinal sections of the finished forgrosion by the "salt-spray" test seems ings. Details of mill practice appear to be largely independent of their to play a role in the production ofchemical composition. So far as the these defects, the occurrence of which authors' tests go, there is no marked in such products as gun forgings was difference in the behavior of several a serious problem during the war. compositions to alternating stress That good manufacturing practice is tests. It appears that the resistance the remedy seems to be the opinion of to the action of alternating stresses many. (See also XVII, Iron and does not increase in proportion as the Steel.) tensile strength of the alloy increases. A study of the origin and develop. (See also Mechanical Engineering, ment of "internal transverse fissures" supra.) in steel rails has led to a promising An important investigation of man-method of investigation. Sections ganese bronze is reported by Merica from failed rails developing transverse and Woodward, "The Behavior of fissures have been etched or "pickled" Wrought Manganese Bronze Exposed to Corrosion while under Tensile Stress" (Proc. Am. Soc. Test. Mats., xix, Pt. II). High-Speed Steel.—Interest in the development of high-speed steels continues, owing to their importance in increased shop production. J. A. Mathews (ibid., Pt. II) describes the development of such steels from the original low-carbon, high-tungsten steel through the introduction of vanadium, the only addition to the earlier types which seems to have afforded universal improvement in quality, down to the recent experiments with cobalt and uranium. Opinions differ as to the value of cobalt; the advantages of "cobaltcrom" steel (A. Y. B., 1918, p. 612) are not fully established. The importance of good steel-making and the need of developing less expensive and laborious tests to determine the efficiency of "red-hard" tools is emphasized. (See also XVII, Iron and Steel.) for two hours in a hot solution of nine parts hydrochloric acid, three parts sulphuric acid, and one part water. The acid opens up transverse and longitudinal cavities the existence of which is not detected by light etching or by the usual methods of tests. Rails free from transverse fissures appear to be free from such cavities. These tests, which are described by F. M. Waring and K. E. Hofammann in a paper entitled "Deep Etching of Rails and Forgings" (Proc. Am. Soc. Test. Mats., xix, Pt. II), have also been applied to new rails and to such forgings as steel wheels and tires. Microscopical and chemical examinations have not so far developed any apparent differences in microstructure that would account for the action of the acid. The authors express the belief that the cause must be sought in some stage of the manufacture. It has been suggested that the fissures that develop in these tests are the result of pressures induced by the forDefects in Steel.-The occurrence mation of hydrogen gas evolved by the of defects, popularly termed "flakes," action of the acid combined with the exhibited in fractures of such alloy internal strain present in the steels steels as nickel and chrome-nickel is tested. The phenomenon appears to discussed by H. S. Rawdon in a be analogous to the cracking of paper on "Microstructural Features of strained brass when attacked by cerFlaky Steel" (Trans. Am. Inst. tain salts (A. Y. B., 1918, p. 611). Mining Engrs., 1919). Macroscopic, Some studies by the U. S. Army Ordincluding X-ray, and microscopic ex-nance Office in connection with the aminations show "flakes" to be coarse manufacture of gun forgings have intercrystalline discontinuities within shown that deep etching reveals de fects that later develop into fissures and into the "flakes" previously referred to. (See also XVII, Iron and Steel.) Magnetic Analysis of Steel.-Magnetic analysis as a determinant of the quality of steel products has continued to receive considerable study. Much of our present knowledge of the subject is reviewed in a symposium held at the 1919 meeting of the American Society for Testing Materials (Proceedings, xix, Pt. II). It is well established that certain magnetic properties are indicative of the mechanical properties of steel, and although much remains to be done along this line, the commercial application of magnetic analysis is most promising. The fact that the magnetic test does not destroy or in any way affect the usefulness of the part tested is of significant importance. The distinct and easily marked changes in magnetic properties resulting from various thermal and mechanical treatments of steel make it possible readily to test such products as cutlery, tools, springs, ball bearings, various automobile parts, etc. Much progress has been made in the detection of flaws, including "transverse fissures" and other inhomogeneities in rails, although there are indications that too great sensitiveness of testing apparatus must be guarded against in order not to cause rejection of material that is satisfactory for all practical purposes. There is reason for confidence in the ability of magnetic analysis to locate faulty rails in service. Another important application is in the study of changes that occur in connection with the phenomenon of fatigue. Phosphorus and Sulphur in Steel.The subject of phosphorus and sulphur in steel continues to be of economic importance. Although the conditions in the raw materials market that led the American Society for Testing Materials in 1918 to increase the phosphorus and sulphur limits in 43 specifications as a war emergency measure (A. Y. B., 1918, p. 613) have been somewhat relieved and the limits restored in 29 of the specifications, the situation is still serious. A comprehensive series of laboratory and service tests is being inaugurated by the Society to throw further light on the effect of these elements and to enable the Society to determine safe limits to include in its various specifications. Corrosion of Steel.-Sheets of uncoated steel of various compositions are being exposed by the American Society for Testing Materials at Annapolis, Md., Pittsburgh, Pa., and Fort Sheridan, Ill., under widely different atmospheric conditions. Semiannual inspections have been made since 1918, but except for some definite failures in the destructive Pittsburgh atmosphere, it is too early to draw conclusions from the tests (Proceedings, xviii, xix, Pt. I). The influence of copper in retarding the corrosion of steel is discussed by D. M. Buck (ibid., xix, Pt. II), who concludes that small percentages materially lower the rate of atmospheric corrosion, steel with 0.03 per cent. copper corroding only 60 to 70 per cent. as rapidly as the same steel with 0.01 per cent. Standard Specifications.-Ten new standards were adopted in 1919 by the American Society for Testing Materials, including specifications for malleable castings, cartridge brass and naval brass rods, tests for flash point of volatile materials, specific gravity of sand and softening point of bituminous materials, and recommended practice for laying sewer pipe (A. S. T. M. Standards Adopted in 1919). Sixty-two standard specifications of the Society have been translated into Spanish by the U. S. Department of Commerce and published as part of an "Industrial Standards" series for distribution among consulate offices in South America as an aid to export trade. Testing and Testing Apparatus.Fatigue of metals is being studied by the Committee on Fatigue Phenomena in Metals of the National Research Council, under the chairmanship of Prof. H. F. Moore. Tests reported in a paper on the subject by Moore and Gehrig (Proc. Am. Soc. Test. Mats., xix, Pt. II) indicate that the static tension test of steel is not a reliable index of fatigue strength; that highstress short-time fatigue tests are not indicative of fatigue strength under oft repeated low stresses; and that certain heat treatment may raise the elastic strength of steel without in creasing its fatigue resistance under operated electrically, upon which the low stresses. The tests also show con- deformations are recorded. The charts clusively that the two methods of of a series of instruments are cencomparing the fatigue resistance of trally driven and marked simultanematerials, namely, first, a comparison ously at definite time intervals. Alof the fibre stresses under which fail- though developed primarily for the ure occurs for a given number of study of concrete ships, the instrument cycles of stress, and, second, a com- promises to be useful in structural parison of the number of cycles re-investigations generally. quired to cause failure under a given stress, may lead to quite different conclusions. Reference should also be made to Moore and Putnam, "Effect of ColdWorking and of Rest on Resistance of Steel to Fatigue under Reversed Stress" (Trans. Am. Inst. Mining Engrs., Feb., 1919). An extensometer for determining the elastic properties of wire, described by T. D. Lynch and P. H. Brace (Proc. Am. Soc. Test. Mats., xix, Pt. II), consists of two flexible side pieces, one end of each being securely fastened to a slotted block, with yokes at their free ends carrying two hardened steel rollers, to each of Tests of electrically welded joints which is fastened a mirror with its in ship plates made by F M. Farmer reflecting surface in the roller axis. in a rotating type of alternating-stress The specimen passes through and is machine indicate that the fatigueresisting strength of the joint, at the present state of the welding art, is about one-twentieth that of the unwelded plate (Proc. Am. Soc. Test. Mats., xix, Pt. II). Comparative impact tests on the Olsen, Izod, and Charpy machines conducted by the Bureau of Aircraft Production (ibid.) indicate that there is no satisfactory method of converting results from one machine into terms of another, but that with proper standardization it is possible to check results on different machines of the same type. The necessity for accurately determining the forces that a ship must resist and the stresses developed under actual service conditions led the Concrete Ship Section of the Emergency Fleet Corporation to devise recording strain gauges which could be placed at many points in the ship and operated simultaneously, so that an exact analysis may be made of the pressures and resulting stresses at any instant. The "strainagraph," as described by F. R. McMillan (Proc. Am. Conc. Inst., xv), consists essentially of a lever system for multiplying the deformation due to stress and a moving chart, clamped to the slotted block and passes between the rollers. As the specimen elongates, the angular movement of the mirrors is magnified by reflecting a beam of light between the mirrors on a curved scale, which is readily calibrated to read to an elongation of 0.00002 in. The instrument combines to an unusual degree the delicacy of the optical lever with a good degree of facility in use. A study of the physical properties of paint by E. C. Bingham and Henry Green (ibid.) has led them to conclude that paint is more properly a plastic material than a viscous liquid, since, unlike a liquid, it will "flow" only under applied pressure, corresponding to the yield point of certain metals. They have developed an instrument, called a "plastometer," which measures the mobility and "yield value" of plastic substances. As applied to paints it determines important properties to which little attention has heretofore been given and may profoundly influence developments in paint technology. Reference should also be made in this connection to Lloyd and Sharples, "Apparatus for Determination of Breaking Point of Pitches” (ibid.). Teaching of Mathematics.-An elo- | It is, therefore, interesting to note quent and forcible answer to those that through an appropriation of $16,who during recent years have at 000 by the General Education Board tacked the position held by mathe- two men, J. W. Young (Dartmouth) matics in our educational curricula and J. A. Foberg (Crane Junior Colwas furnished by the very large lege), have been enabled to devote. amount of mathematical teaching de- their entire time for a year, beginmanded by the Students Army Train- ning July 1, 1919, to a study of the ing Corps. The suggestive mathemat- general mathematics-teaching situaical course prepared for these stud- tion, collaborating with the Commitents by a volunteer committee of tee on Mathematical Requirements teachers was fully considered during of the Mathematical Association of the Dartmouth meeting of the Mathe- America. Since so little has been matical Association of America. published on college mathematics The greatest mathematical war teaching, it is of interest note here service was rendered by the thousands the appearance of Paul Klapper's of teachers who gave mathematical book on College Teaching (World instruction in the universities and in Book Co.), which contains a chapter the various Army and Navy camps at on the teaching of college mathemathome and abroad. Many teachers of ics. other subjects volunteered to assume the unusual burden of teaching a subject which they had largely forgotten in view of the inadequate supply of trained teachers of mathematics. The Y. M. C. A. rendered conspicuous service in its "Khaki University," especially in the development of brief but serviceable mathematical courses. J. W. Young (Dartmouth) acted as general director, and E. R. Hedrich (Missouri) was in charge of this service among the American soldiers in France (see XI, The Army; and XXX, Education). Mathematical Research.-Although as a means of winning the Great War, the need of good mathematical teaching was more widely recognized than that of good mathematical research, yet the need of the latter did not remain unnoticed. According to a circular letter issued in June, 1919, by the chief of the technical staff of the Ordnance Department of the Army, there was the most gratifying result of substantial progress and the opening of new fields in which it is essential that the Ordnance Department continue investigation. change in the form of the projectile for will serve Notwithstanding the importance of For example, it was found that the slight good mathematical teaching, little the 6" gun resulted in the range being has been done towards a systematic increased two and one-half miles. study of the entire subject with a highest grade of men, expert in matheIt is therefore important to secure the view to suggesting possible improve- matics and dynamics, who ments. The International Commis- their country in the development of ordsion on the Teaching of Mathematics, nance material to the greatest possible degree of superiority. with its numerous subcommittees, made a noteworthy beginning in this direction, but the men who prepared the reports for these committees had to do the work during whatever time they could spare from other duties. That the needs of the war have made a strong impression on mathematical research in this country is evident from various papers read before the American Mathematical So |