and a summary of their replies prepared; but as the hard-rubber division of the Rubber Association has promised to submit_recommendations, the completion of the specifications has been deferred until these recommendations shall have been received.

Construction of Storage Batteries.-Work on this subject has been confined largely to a study of the oxides of lead used in storagebattery manufacture and to the preparation of pastes from these oxides. As no recognized standard method for determining the acid-absorption numbers of these oxides is available, several possible methods have been studied and compared; and the effects produced by varying the concentration of the acid and the duration of the experiment have been studied.

A study of the properties of pastes shows that the consistency of the paste, the strength of solution, the kind of oxide, the humidity, the temperature, and the time consumed in the processes all affect the characteristics of the finished plate. A method for measuring the consistency and the rate of setting of the paste is needed. Some experiments relative to this have been made, and an empirical method for measuring consistency has been devised and followed in all of the later work. Experiments have been made on four special samples of oxides, covering a large range in size of particles and in red-lead content. In order to obtain a paste of the proper consistency, larger volumes of solution must be added to the oxides containing the smaller percentages of red lead. Pastes having a high percentage of red lead set more quickly than others.

A large number of experimental plates has been pasted and assembled into cells for tests in the laboratory. Expanders for negative plates have been made with various percentages of lamp black and barium sulphate. A study has also been made of the effect of the treatment of plates after being pasted, including air drying, dipping, and pickling processes.

Methods of Magnetic Testing. There are several types of magnetic permeameters in commercial use, and a satisfactory method for standardizing them is needed. A study of various possible testing methods leads to the conclusion that the one involving the use of the Burrows permeameter, developed at the bureau, is the most satisfactory basic method when standard specimens of the requisite degree of magnetic uniformity can be obtained. However, it is difficult to find such specimens. Therefore a series of specimens of specially selected and specially heat-treated steel has been prepared; a few of them have been found to be suitable for the purpose. Details of the work are reported in a paper to appear in the Bureau of Standards Journal of Research.

A standard method of making magnetic measurements with intense fields is needed for the testing of the newer types of permanentmagnet steels. This has been sought, but as yet no satisfactory standard method has been found. Several methods that have been developed for commercial use have been intercompared, and the sources of possible error studied.

Of the well-known methods for measuring, with alternating currents and at low inductions, the permeability and core loss of materials used for magnetic cores, none appears to be entirely satisfactory. New methods are being developed in cooperation with industrial laboratories.

Magnetic Properties. For the proper development of magnetic analysis, fundamental data on the relationships between magnetic and other physical properties must be obtained; investigations in this field have been continued. These have included the relation between the magnetic properties and the torsion characteristics and between the results of impact tests and the magnetic properties of a high-carbon drill steel; both studies were carried on in cooperation with other laboratories. The installation and adjustment of apparatus for studying thermal effects have been completed, and a description of the apparatus has been published.

In a completed investigation of the effect of repeated stress on the magnetic properties of steel, it was found that although changes in magnetic properties accompany the progress of a fatigue test, there is no apparent relation between these changes and the progress of the fatigue failure. The observed changes in magnetic properties are attributed to changes in stress conditions within the specimen; these stresses complicate the relation between magnetic and other physical properties. The results of this investigation have been published. An apparatus has been set up for measuring the magnetic properties of wires while subjected to mechanical tension, or to high temperature, or simultaneously to both. In this work attention will first be given to the properties of approximately pure iron.

Publications. Apparatus for Thermomagnetic Analysis, R. L. Sanford, B. S. Jour. of Research, 2, No. 4, p. 659, April, 1929.

Note on the Effect of Repeated Stresses on the Magnetic Properties of Steel, M. F. Fischer, B. S. Jour. of Research, 1, No. 5, p. 721, November, 1928.

EMPLOYMENT AND FUBLIC WORKS

In cooperation with the National Bureau of Economic Research, the bureau is studying the extent to which the construction of Federal, State, and local public works can be so adjusted as to afford some practical relief from unemployment during periods of depression, and thus to contribute to general business stability. Such an adjustment will assist in stabilizing building activities, and will be very pertinent to city and regional planning, which involve long-term financial budgets.

FIRE RESISTANCE

Fire Resistance of Walls.-Comprehensive tests to determine the fire resistance of hollow clay wall tile, conducted in cooperation with the Structural Clay Tile Association, have been completed and the results published. The main series consisted of 167 fireendurance tests, and 4 fire-and-water tests of typical wall constructions; 71 were made with walls from 10 to 11 feet high, and 8 to 16 feet wide. This was preceded by two preliminary series of tests made to determine the effects of fire on individual tile units, and the effect of changes in design of the unit and in the constituents and the preparation of the clay.

The results are summarized in terms of fire-resistance periods. This period is determined by the shortest of the following intervals: (1) The time the wall sustained its load under fire exposure, (2) the time required for the average temperature on the unexposed side to rise 130° C. (250° F.), and (3) the time required for the temperature to reach 180.6° C. (325° F.) at any point where temperature measure

ments are taken. Freedom from cracks and openings large enough to transmit flame or ignite combustible materials is also required both in fire-endurance tests and in fire-and-water tests.

Fire tests have also been conducted on sand-lime and concrete brick walls that had been seasoned at temperatures considerably above room temperature so as to drive off the excess of water introduced into the brick during manufacture.

Publications.-Fire Resistance of Hollow Load-Bearing Wall Tile, S. H. Ingberg and H. D. Foster, B. S. Jour. of Research, 2, No. 1, p. 1, January, 1929.

Fire Resistance Tests with Particular Reference to Wall Constructions, S. H. Ingberg, Proceedings Twenty-ninth Annual Convention of the Sand-Lime Brick Association, p. 37, 1929.

Severity of Building Fires.-A burning-out test was conducted of a fire-resistive building wherein kerosene, contained in open pans, was introduced to an amount equivalent in British thermal units to the furniture and paper that were contained in the same building with office occupancy previously tested. Although the weight of the kerosene introduced was less than one-half of that of the furniture and paper in the companion test, the equivalent duration of the fire was the same within 10 minutes. The results indicate that the amount and the fuel value of the combustibles present per unit of floor area or of room volume may possibly be used as a means for gaging the fire hazard of an occupancy in terms of the severity of the standard furnace test.

Publication.-The Severity of Fires in Buildings, by S. H. Ingberg, The Architectural Forum, 1, No. 5, p. 37, May, 1929.

Fire Hazard of Jute.-During the past nine years there have been reported 125 fires in jute cargoes in transit between India and American ports or in jute warehouses, most of which have been attributed to spontaneous ignition of the jute. A research on the fire hazard involved is being conducted jointly by the Bureau of Chemistry and Soils of the Department of Agriculture and the Bureau of Standards, in cooperation with the Marine Underwriters, which organization has appointed a research associate to work with the two bureaus on the problem. Previous tests have indicated that typical Indian jute does not present any decided susceptibility to spontaneous heating from oxidation effects alone. Some heating from bacteriological action has been found to occur, resulting in the so-called heart damage. The work is therefore planned to determine the effects from both bacteria and oxidation. Large heating chambers that will hold a whole bale of jute have been built and equipped with control for temperatures up to 250° C. Means are also provided for humidity control and for sampling and analyzing the gases evolved. The moisture content and bacterial flora of the bales will be varied in efforts to obtain conditions most favorable for heating.

FOUNDRY PRACTICE

Foundry Sands. For several years the bureau has been cooperating with the American Foundrymen's Association in the development of standard methods for testing and grading foundry sands. Descriptions of these methods are now available from the association.

Core Binders.-The bureau has continued its work on the use of rubber and allied materials as core binders in foundry practice. The

use of such binders obviates the baking of the cores. A widespread interest in binders of this type is shown by the number of requests that have been received for Letter Circular No. 252 giving information regarding them.

Suspensions of rubber latex in water have recently been used with pronounced success. These are free from the fire hazard that attends the use of rubber dissolved in inflammable solvents.

Publication.-B. S. Letter Circular No. 252.

GAS ANALYSIS AND SERVICE

Fractional Distillation of Gases.-An extended study of methods for the analytical separation of natural gases by fractional distillation, including the definition and determination of the gasoline in the gas, has been completed. Both the definition of "gasoline" and a reference method for determining its amount are of primary importance in the sale of this important commodity.

Publications.-Accurate Determination of the Gasoline Content of Natural Gas and the Analytical Separation of Natural Gases by Isothermal Fractional Distillation, G. M. Shepherd, B. S. Jour. of Research, 2, No. 6, p. 1145, June, 1929. Exact Determination of the Gasoline Content of Natural Gas, G. M. Shepherd, Oil and Gas Journal, 26, p. 48, 1929.

An Automatic Sample-Collecting Vacuum Pump, E. R. Weaver and Martin Shepherd, Jour. Am. Chem. Soc., 50, p. 1829, 1928.

New Gas Analysis Method for Natural Gasoline Industry Is Perfected, G. M. Shepherd, National Petroleum News, 21, p. 31, 1929.

Orifice Meters.-The results of the experiments on compressed air carried out at Edgewood Arsenal in cooperation with the Chemical Warefare Service have been published; and those obtained at Chicago in cooperation with the American Gas Association are being prepared for publication. The bureau's cooperation in carrying out the comprehensive program of investigation outlined by a committee of the natural gas department of the association is being continued and extended.

Balances for Specific Gravity and Density. Following a study of the relative merits of available balances for determining the specific gravity of gases, and of the conditions of field use of such equipment, an improved balance was constructed, employing a type of suspension not previously used in such instruments. A balance has also been designed, but not yet constructed, for the direct determination of the density of gases. These instruments were developed in cooperation with the Natural Gasoline Association of America. The determination of the density is an important factor in measuring natural gas for sale. Burners for Propane and for Butane.-Extensive work has been done on the design of domestic burners for the gases propane and butane, which have only recently been used extensively. In determining the optimum construction, consideration was given to the size of ports and orifices, to the distance between burners and utensils, to the gas pressure, to the amount of gas used, to the time required for heating utensils, and to safety from the principal hazards involved in the use of gas ranges. Successful laboratory burners for these "bottled" gases were also constructed, and are now being used as models by apparatus companies.

Oven Linings. Apparatus has been constructed with which it is possible to test quickly the corrosion of oven linings under conditions

which are believed to reproduce the relative rates of corrosion in service. The majority of available commercial linings have been tested. The method was developed in cooperation with the American Gas Association, for one of its principal objects was the determination of a suitable standard of resistance to corrosion, a standard that should be met by appliances receiving the association's approval.

Miscellaneous Gas Appliances.-Much time has been given to a study of storage water heaters as regards efficiency, methods of rating, and satisfactory performance standards. Critical studies made of several novel burners, appliances, and accessories for the use of gases of various sorts have led to the elimination from the market of at least three dangerous or unsatisfactory pieces of equipment, and to a substantial improvement of several others. A study has been made of the floor temperatures under radiant heaters so that standard safety tests may be devised.

Publications.-Idoine Pentoxide Method for Analyzing Products of Combustion for Small Quantities of Carbon Monoxide, J. H. Eiseman, Gas Chemists' Handbook, third edition, p. 289.

Application of the Thermal Conductivity Method of Gas Analysis to the Study of Gas Appliances, S. L. Earle, Gas Chemists' Handbook, third edition, p. 297.

GLASS AND VITREOUS ENAMELS

Production of Optical Glass.-Thirty-nine pots of optical glass were made in the course of the bureau's research to determine melting procedures which will consistently yield optical glass of good quality. Six different kinds of glass were made, borosilicate crown, barium flint, medium flint, dense flint, light barium crown, and ordinary crown. Twenty-five thousand nine hundred and ten blanks of various sizes and shapes, weighing approximately 3,500 pounds in all, were molded and annealed. The majority of these blanks were of excellent quality, and were used by the Navy Department.

Viscosity of Glass.-A study of the viscosity-temperature relations of light barium crown and of borosilicate crown glasses melted in the bureau's glass plant has been made in the hope that these relations will indicate how the light barium crown should be melted in order to produce glass of a quality comparable with that obtained in the borosilicate crown glass.

These relations can be expressed by the equation:

in which:

Log. μ=C- 13 √b2 — (1400 – T2)

B

μ = viscosity, T=temperature (°C.)

and A, B, and C are constants having the following values for the two types of glasses:

Effect of Thermal Treatment.-Any change in the annealing schedule that causes a change in the effective temperature of annealing will produce changes in certain properties of the glass. Additional