various industrial plants and laboratories, affording reliable means of checking the accuracy of industrial analyses. The Bureau also prepares and maintains the standards of combustion, used in standardizing calorimeters, consisting of highly purified materials of constant heating values, such as naphthalene and benzoic acid. The distribution of nearly 100 standard combustion samples during the year has already done much to place the testing of fuels upon a uniform and comparable basis.

COMPARISON OF THE STANDARDS.

The comparison of standards has been carried on for the general public, including those engaged in engineering, manufacturing, and commerce, as well as those engaged in scientific and technical work. Many comparisons of length standards were made during the year for municipal and State governments. The capacity standards tested included those for Porto Rico, a set for the Panama Railroad, and sets for the States of Wisconsin and Alabama.

The purpose of comparing standards is to enable the manufacturers. to make instruments of the highest grade, according to the most. approved specifications. By testing the manufacturers' standards the public is assured of accuracy, since by this means precision is established at the outset, namely, in the factory. It is also possible for the manufacturers and the public to have such instruments retested from time to time to check their accuracy. The Bureau is constantly engaged in intercomparing its standards with those of standardizing institutions of other nations, since international trade demands that the units of measure shall be to the highest degree uniform throughout the world. Only recently have such comparisons given promise of a substantial uniformity.

During the year, by act of Congress, the old troy pound of the Mint was superseded by the standard of the Bureau. This places all measures of mass in the United States upon the basis of the standard kilogram.

CONSTRUCTION OF STANDARDS.

The Bureau now has excellent facilities for the construction of standards. A corps of expert mechanicians, with a well-equipped shop, are available for such work, which may be done under the direct supervision of the experts who design the standards and instruments. New standards are based upon careful study of the best practice, and in some cases the construction is prescribed by international congresses. Multiples and subdivisions of the standards are constructed to provide for all ranges of measurement. Their values must be established by comparison with the primary standards. During the year four separate standards of electrical resistance, con26321° -C & L 1911

6

structed at the Bureau, agreed within 5 parts in 1,000,000—a highly gratifying degree of precision.

On January 1 the Bureau adopted, by international agreement, the new value for the international volt. This creates uniformity between England, France, Germany, and other countries.

During the year a number of new photometric standards were prepared and studied as to their reproducibility and constancy. Carbon-lamp standards have been prepared as standards of light and intensity and distributed to the various departments of the Govern

ment.

TESTING AND CALIBRATION OF STANDARDS AND INSTRUMENTS.

The testing and calibration of standard measuring apparatus during the year included a wide variety-length measures, from microscopic scales to geodetic tapes; balances; sealers' measures; chemical measuring apparatus; standard instruments for temperature measurements, adapted to ranges from liquid hydrogen to the highest attainable temperatures, flash-point apparatus, etc.; standards of composition for materials; standards of light intensity, such as electric lamps, gas-flame standards, and many others.

The testing of such standards and measuring apparatus is in many cases for the purpose of insuring compliance with contract specifications. In other cases the tests are made because the required precision is not attainable elsewhere, and in still more important cases the manufacturers themselves are establishing, by means of the Bureau, standards of measurement for their industries. By cooperation with the manufacturers, for example, the thermometer industry and the lighting industry have been placed upon an industrial uniformity heretofore impossible.

There is being published the results of an extended investigation of electric switchboard instruments. This was undertaken at the request of the Navy Department, and the results will be made public with the approval of the manufacturers concerned.

During the year a comparative test has been made of various systems of master clocks and their secondaries. The test covered the leading makes and continued for six months. The results were placed at the disposal of the Treasury Department for use in selecting the most efficient system for Government buildings.

As to the extent of this work, 853 electrical standards and measuring instruments have been tested; nearly 19,000 thermometers were tested during the year, besides tests of 90 high-temperature measuring instruments and materials; 240 standards of length and length-measuring instruments were compared; 2,400 hydrometers, used to determine the density of liquids, were tested, largely for the Internal-Revenue Service; 9,000 volumetric measures were tested

during the year, including cubic-foot bottles for gas-meter testing, sealers' standards, and chemical measuring apparatus; and the Bureau's traveling inspectors visited the factories and inspected 944,000 electric lamps for the Government; of these, 3,483 were tested by the Bureau for life and candlepower. In addition, about 300 photometric standards of various kinds were prepared.

PROBLEMS IN CONNECTION WITH STANDARDS.

In connection with standards of every kind, many scientific problems arise which must be solved in order to improve the standards or the methods of using them. Thus, progress in standardization depends upon the results of such researches. Several examples of such problems completed or in progress during the last year may be mentioned.

Important progress has been made in the designing of apparatus to regulate temperatures in length measurements, resulting in an excellent temperature control-a condition vital to precision work. For heat measurements new comparators, furnaces, and testing equipment have been designed. Two investigations in progress during the year have had for their object the more rigorous determination of the unit of electric current. The results of the investigations have been prepared for publication.

The Bureau is cooperating with the national and international technical societies wherever such cooperation gives promise of furthering the standardization of methods and specifications. Among the cases may be cited the cooperation with the American Chemical Society as to standards of composition and purity of materials; with the American Society for Testing Materials as to methods of testing, standards, specifications, and other subjects; with the American Institute of Electrical Engineers in the establishment of standardized definitions and practice. As a result of careful determination of conductivity of commercial copper, the Bureau prepared tables of copper-wire constants. These copper-wire tables have been approved by the latter society and will soon be published as a suggested basis for international agreement.

In establishing standards of gaslight intensity much work has been done on the Harcourt Pentane standard to secure reliable results. The conditions affecting constancy are being ascertained and a great improvement in gaslight measurements may be expected. Other problems include the study of the sources of error in aneroid barometers for the purpose of adopting standard methods of testing; a research to test the value of purified benzoic acid as a standard of acidity in acidimetry; the refinement of sodium sulphate which at a definite temperature exhibits a marked change in conditions well

adapted for use as a reliable fixed point in thermometry; a research aiming to free naphthalene from obscured impurities.

The problem of producing pure oxygen in large quantities has now been solved by newly devised apparatus. The oxygen is used for combustion purposes in precision testing of heat values. A number of highly purified gases for heat determination have also been prepared as a basis for gas engineering tables. Such tables will permit the calculation of the heat value of illuminating gas direct from the chemical analysis.

During the year there were published the results of 44 investigations concerning standards, measuring instruments, and methods of

measurement.

DETERMINATION OF PHYSICAL CONSTANTS.

Physical constants are the measured data relating to materials and energy which underlie scientific and technical work. Hence they are of fundamental importance in the work of standardization. Among these would be the melting and boiling points of materials, flash points, breaking and crushing strength, hardness, density, expansivity, and conductivity. A few researches of this character may be cited, e. g., the thermal expansion of 25 different specimens was determined for ranges from below freezing to above the boiling point of water; the density and thermal expansion of linseed oil and turpentine was determined, a constant much used in specifications for purity.

In connection with standards for commercial use, the density of ethyl alcohol is of great importance. This physical constant has been under investigation at the Bureau for several years. The alcohol was purified by the most refined methods which could be devised and the density was determined with high accuracy, probably correct to within 1 part in 100,000. The values of the densities of alcohol-and-water mixtures were determined with an accuracy within 2 parts in 100,000.

A physical constant of importance in atomic-weight work is the atomic weight of bromine, which is now being determined by the Bureau as a logical conclusion to similar work on chlorine. The materials have been purified and the work is well advanced.

A number of boiling-point and freezing-point determinations have been made for use in temperature work. For example, the boiling point of sulphur has been accurately determined for use as a fixed point in thermometry corresponding to the temperature 444.7 C. The concordance of this work gives promise of international agreement on the scale of temperature above the boiling point of water, ranges in which uncertainty has always existed.

In practically every line of work similar researches are in progress, with the aim of providing reliable data upon which to base units and standards. The pressing need for such work and its utility can hardly be overestimated.

DETERMINATION OF THE PROPERTIES OF MATERIALS.

The determination of the quality of materials is a most important problem in the industries, since to the consumer quality is the real object of purchase and sale. The modern view that value rests upon measurable properties which combine to fix the quality is the basis for the determination of quality by measurement.

The materials tested by the Bureau during the year include such materials as the cements, metals, and other materials used in the construction of the Panama Canal; the supplies for the executive departments at Washington; and bookmaking materials-paper, type, ink, cloth, and glue-used by the Public Printer for Government documents. Here, as elsewhere, the investigations have resulted in great saving and efficiency in buying by basing such purchases upon tested quality and price alone.

Of the determinations of the properties of materials involving chemical work there were during the year 8,868 separate tests for practically every department of the Government and many times. this number of separate determinations. Important investigations are in progress in connection with the properties of paper, paper pulps, and the composition and manufacture of paper. The properties of the new monel metal are being investigated to determine its suitability for laboratory utensils and appliances. The properties of materials—such as steels, irons, refractory materials, and other materials of construction-have been under investigation to determine their suitability for the particular uses for which they are intended. The purity of platinum sold for chemical purposes has been studied with the greatest care to ascertain the nature of the impurities present a research of vital importance to chemical analysts. In this work the Bureau has cooperated with other laboratories concerned. The Bureau is cooperating, also, in an international comparison of the magnetic properties of materials.

TRADE WEIGHTS AND MEASURES INVESTIGATION.

The investigation of trade weights and measures conditions throughout the country, for which Congress has made special provision, has practically been completed and will form the subject of a special report. Every State in the Union, with the exception of Arkansas and Oklahoma, has been visited, and inspections in the two mentioned will be made in the near future.