ments shall be admitted to official verification, to adopt regulations concerning the material, construction, or designation of the apparatus, to regulate the methods employed in the testing and verification, and to fix the fees and specify the seal to be employed.

SEC. 11. Measuring apparatus verified in accordance with this law may be used in trade in any part of the Empire.

SEC. 12. Whoever, engaged in the industrial supply of electrical energy, does not comply with section 6, or the regulations based thereon, will be subjected to a fine not exceeding 100 marks, or imprisonment not to exceed four weeks. In addition, the incorrect instruments, or the instruments not complying with the regulations, shall be subject to seizure.

SEC. 13. This law and the regulations adopted in accordance with sections 6 and 12 shall take effect January 1, 1902; the remainder on the date of its promulgation.

REGULATIONS FOR CARRYING OUT THE LAW ISSUED BY THE BUNDESRATH.

[Reichsgesetzblatt No. 16, 1901, June 1, 1898.]

In accordance with paragraph 5 of the law defining the electrical units of measurement (Reichsgesetzblatt, p. 905), the following specifications are adopted:

1. Conditions under which the silver is to be deposited in the specification of the ampere. (Sec. 5a). The solution shall consist of from 20 to 40 parts by weight of pure silver nitrate in 100 parts of distilled water free from chlorine. It shall not be used after 3 grammes of silver are deposited from 100 cubic centimeters of solution.

All parts of the anode in contact with the solution shall consist of pure silver. The cathode shall consist of platinum. When the deposited silver exceeds 0.1 gramme per square centimeter the silver is to be removed.

The current density at the anode shall not exceed one-fifth, and at the cathode one-fiftieth ampere per square centimeter.

Before weighing, the cathode is first to be rinsed with distilled water free from chlorine until the addition of a drop of hydrochloric acid to the wash water produces no opalescence. The cathode is then to be soaked for ten minutes in distilled water at 70° to 90° C., and is to be finally washed with distilled water. The last wash water after cooling must not become opalescent upon the addition of hydrochloric acid. The cathode is dried by the aid of heat and kept in a desiccator until it is weighed, which shall not be done less than ten minutes after cooling off.

2. Designation of the electrical units (sec. 5b).—(a) The quantity of electricity flowing through the cross section of a conductor in one second when the current in the same is equal to 1 ampere is called an ampere-second (coulomb), and the quantity flowing in one hour an ampere hour.

(b) The power corresponding to an ampere in a conductor having a potential difference of 1 volt between its terminals is called a watt.

(c) The work done in one hour when the power is equal to 1 watt is called a watt hour.

(d) The capacity of a condenser which is charged by an ampere-second to 1 volt is called a farad.

(e) The self-inductance of a conductor in which 1 volt is induced by a uniform change in the current of 1 ampere per second is called a henry.

3. Designations for the multiples and submultiples of the electrical units (sec. 5c).—The following prefixes to the name of a unit shall have the following meanings:

Kilo.....

Mega (meg).
Milli

Micro (mikr)..

EXPLANATION OF THE SPECIFICATIONS.

1,000 times 1,000,000 times One one-thousandth

One-millionth

(SEC. 5a.)

1. On account of its fundamental importance in defining unit current, the silver voltameter must be employed under prescribed conditions in order that the amount of silver deposited will give the correct value for the current measured. In this connection the fact is to be considered that the silver solution gradually experiences a yet unexplained change by the long-continued passage of an electric current, by which small variations in the amount of silver deposited are produced. Moreover, in order to obtain compact deposits of silver the concentration of the solution and the current density at the electrodes must be confined within certain limits. Finally, on account of the sensitiveness of the silver solution to impurities it appeared necessary to prescribe regulations concerning the water employed in washing, and in order to weigh only the deposited silver to specify the manner of washing.

In these specifications the experiences of the most reliable experimenters have been considered. They may be considered not only entirely sufficient for their purpose, but in addition they are possibly somewhat too rigorous for practical needs. Since, however, it is really not necessary, except in rare cases, to refer current measurements to the silver voltameter, as much simpler methods are available, the specifications could be rigorously drawn.

LEGAL DEFINITION OF THE ELECTRICAL UNITS IN

AUSTRIA.

[Ordinance No. 176, Ministry of Commerce, of July 4, 1900. Concerning the testing and certifying of electrical supply meters.]

The electrical units are derived from the fundamental metrical units of length, mass, and time, according to the electro-magnetic system of measurement, taking the centimeter as the unit of length, the gramme as the unit of mass, and the mean solar second, of which there are 86,400 in a mean solar day, as the unit of time. The resulting units are designated as units of the c. g. s. electro-magnetic system (centimeter-gramme-second system).

The unit of resistance is the ohm, which is equal to 109 units of resistance of the electro-magnetic c. g. s. system. For commercial purposes the Ohm. ohm may be considered equal to the resistance offered to an unvarying current by a column of mercury having a mass of 14.4521 grammes, a length of 106.3 centimeters, at the temperature of melting ice. (No reference is made to a uniform cross section of the mercurial column.)

The unit of current is the ampere, which is equal to the one-tenth part of the electro-magnetic unit of current of the c. g. s. system. For comAmpere. mercial purposes the ampere may be considered equal to the value of an unvarying current which when passing through an aqueous solution of silver nitrate deposits 0.001118 gramme silver per second.

The unit of electromotive force is the volt, which is equal to that electromotive force which acting steadily on a conductor having a resistance of 1 ohm produces in the same a current of 1 ampere.

Volt.

Watt.

The unit of power is the watt, which is equal to 107 units of power of the c. g. s. system, or equal to the power corresponding to a current of 1 ampere at an electromotive force of 1 volt (voltampere). The coulomb is equal to the quantity of electricity flowing in one mean solar second through a conductor carrying a current of 1 ampere. One ampere-hour corresponds to 3,600 coulombs.

Coulomb.

The work done in 3,600 seconds in a conductor in which the power is 1 watt is equal to 1 watthour. One hundred watthours are equal to 1 hectowatthour. One thousand watthours are equal to 1 kilowatthour.

Watthour.

LEGAL DEFINITION OF THE ELECTRICAL UNITS IN

FRANCE.

[Decree of the President, April 25, 1896.]

The unit of electrical resistance the ohm-is the resistance offered to an unvary ing current by a column of mercury at a temperature of melting Ohm. ice, having a mass of 14.4521 grammes, a constant cross section,

and a length of 106.3 centimeters. The unit of current-the ampere-is the one-tenth of the electro-magnetic unit of current. It is represented sufficiently well for practical purposes Ampere. by the unvarying current which deposits in one second 0.001118

grammes of silver. The unit of electro-motive force-the volt—is the electro-motive force which produces a current of 1 ampere in a conductor having a resistance of 1 ohm. It is represented sufficiently well for practical purposes by 0.6974, or (1989) of the electro-motive force of the Latimer Clark cell.

Volt.

SPECIFICATIONS FOR THE SILVER VOLTAMETER.

The specifications for the silver voltameter are the same as those adopted by the Chicago congress, with the exception that the anode is to be supported by a silver rod instead of platinum wires, the latter having been found to be a source of variations. In addition, the directions for making a measurement are the same as those given in the English law.

SPECIFICATIONS FOR THE STANDARD CELL.

The specifications for the standard cell are the Board of Trade specifications, which are legalized in England.

PROPOSED LAW DEFINING THE ELECTRICAL UNITS FOR USE IN BELGIUM.

There is established for the Kingdom a single system of electrical units having as its base the ohm, the ampere, and the volt.

Ohm.

The ohm is the resistance offered to an unvarying current, by a column of mercury at the temperature of melting ice, having a mass of 14.4521 grammes, a constant cross section, and a length of 106.3 centimeters. The ampere is represented sufficiently well for practical purposes by the intensity of a constant current which precipitates in one second 0.001118 Ampere. grammes of silver from an aqueous solution of silver nitrate.

Volt.

The volt is represented by the electro-motive force which produces a current of 1 ampere in a conductor of which the resistance is 1 ohm.

The denominations of the derived electrical units, especially the units of energy and power, shall be fixed by royal decree. Multiples and submultiples of the legal units may likewise be fixed.

Derived units.

Within two years after the promulgation of the law practical standards in conformity with the legal system of units shall be established by a special commission named by the King.

The custody and periodic verification of these standards shall be entrusted to the minister of industry and labor.

PROPOSED LAW DEFINING THE ELECTRICAL UNITS IN SWITZERLAND.

The law proposed by the Swiss Electro-Technische Verein, and drawn up by the president of the commission on inspection service and units of measurement, is substantially the same as the German law.

OTHER COUNTRIES.

In Mexico no electrical units have been legalized, but a bill is under consideration defining the electrical units, which at the suggestion of the Bureau of Standards will be so worded that any modifications adopted at St. Louis may be made.

In Norway, Sweden, Denmark, Netherlands, Portugal, Italy, Japan, and Russia no units have been legalized.

No definite information has been obtained from Spain.

THE SPECTRA OF MIXED GASES.

By P. G. NUTTING.

Early in spectroscopic work it was observed that mixed gases frequently gave the spectrum of a heavy metallic component more strongly than that of a lighter component. In 1878 E. Wiedemann" described experiments with mercury and sodium in hydrogen and nitrogen. He proved that the intensities of the metallic spectra were out of all proportion to the relative amounts of metallic vapor present. It is a matter of common experience in preparing Plücker tubes to exhibit metallic spectra that as soon as the tube is heated sufficiently to vaporize the contained metal, the spectrum of the lighter gas filling the tube disappears. Recently Professor Lewis has shown that when mercury vapor is present in a tube of hydrogen it will reduce the hydrogen spectrum to about half its original intensity when but one molecule of mercury to three thousand of hydrogen are present. It is well known that the cadmium spectrum will swamp that of hydrogen long before the cadmium is even melted. The red lines of the two spectra are equal at a temperature of about 200° C.

At first thought we should say in explanation that the vapor of a metal would be a better electrical conductor than that of a nonmetal, and, carrying the most of the current, would show the brightest spectrum. But this hypothesis is clearly untenable. Although gas conductivities have not yet been accurately determined nor yet even defined, it is well known that under the same conditions metallic vapors do not differ widely from nonmetallic vapors and the permanent gases in conducting power. And even then, why should the vast majority of lighter molecules be left idle as soon as a few of the heavier molecules are present?

Taking up the problem at this point, it was quickly shown that metallic character has little if anything to do with spectral predominance. Sulphur and iodine are nearly as effective in swamping

a E. Wiedemann: Wied. Ann., 5, pp. 500-524; 1878.

P. Lewis: Astroph. J., 10, pp. 137-163; 1899. Ann. d. Phys., 2, pp. 447-458; 1900.