Lapas attēli

Repeated object lessons have demonstrated that nearly all progress in science has resulted in important advances in industry


G-E Research Laboratory

Schenectady, N. Y.

Among the many products developed by the General
Electric Company's research laboratories the following
are of special interest to manufacturers:

[blocks in formation]

For further information address Supply Department, Schenectady Office,


General Electric

General Office
Schenectady, N.Y.

Sales Offices in all large cities


[merged small][merged small][merged small][merged small][ocr errors][merged small]

Interference Refractometer, Rayleigh-Hilger for Gases. This instrument is designed for work on the refractometry of gases, of which the refractive indices differ so little from unity that a very high degree of sensitivity is required in any refractometer to be used with them. Some idea will be obtained as to the sensitivity of the instrument when it is stated that it is possible by means of it to detect the presence of .01% of hydrogen in air, a quantity which causes a change of refractive index of only .000,000,015. The principle underlying the design of the instrument is that of the interference bands formed by two adjacent slits in collimated light, and observed by means of a telescope. The arrangement is, in fact, a diffraction grating with only two openings. The means whereby the system of bands thus obtained is made to indicate the refractive index of a substance is provided by the sideway shift of the whole band system when any change is made in the optical path of one of the interfering beams; part of the optical path being, of course, the substance under examination.

Quantities of Gases in Air Detectable by Rayleigh Interference Refractometer Minimum observable shift assumed to be 1/40 band. Hydrogen. .01% Helium .006% Chlorine.

.003% Nitrogen.. .3% Sulphur Dioxide. .034% Carbon Monoxide

.03% Nitric Oxide. .0065% Water..

.04% Carbon Dioxide. .0095% Argon.. .15% Hydrogen Sulphide. .0045% Ammonia...

.017% Bibliography of Rayleigh Interference Refractometer, Original Arrangement and Laboratory Type Lord Rayleigh.

Proc. Roy. Soc. (A) vol. 59, p. 201 (1896).
Proc. Roy. Soc. (A) vol. 64, p. 97 (1898).

Proc. Roy. Soc. (A) vol. 62, p. 225 (1897). Ramsay & Travers.

Proc. Roy. Soc. (A) vol. 64, p. 190 (1899).

Proc. Roy. Soc. (A) vol. 67, p. 331 (1900). Travers.

Study of Gases," Macmillan, 1901. J. D. Edwards..

Technologic Papers of the Bureau of Standards. {

No. 113.

( Jour. Amer. Chem. Soc. 89, p. 2382, 1917. Adams...

Jour. Amer. Chem. Soc. 39, p. 1181, 1915. 45940. Refractometer, Interference, Rayleigh-Hilger, for gases, with double gas tube 100 cm. long, but without illuminating lamp..

..654.75 NOTE-By, the substitution of a Liquid Cell in place of the Gas Tube, the Rayleigh Interference Refractometer can be converted into an instrument for the investigation of liquids, or can be furnished for this purpose without the Gas Tube. Prices on application.

Prices subject to change without notice





MAGNETISM1 The Electron Theory of Magnetism: DR. 8.

EVER since the time of Faraday it has been J, BARNETT

465 known that all varieties of matter can be

grouped in three classes on the basis of magThe American Association for the Advancement of Soience:

netic behavior, ferromagnetic, paramagnetic

and diamagnetic. Spring Meeting of the Executive Committee: DR. BURTON E. LIVINGSTON...


It would be far too much to claim that the

electron theory has as yet given anything like Medals of the National Academy of Sciences :

a complete account of the phenomena witDR. E. E. SLOSSON,


nessed in connection with these three types of Third Award of the Daniel Giraud Elliot magnetism; but it is the only theory proposed Medal: DR. HENRY FAIRFIELD OSBORN.... 480 which has been in any way satisfactory and

which appears to hold out any hope for the Scientific Events:

future. The United States Patent Office; The National Geographic Society; Exchange of

In accordance with the plans of this symProfessors of Engineering between Ameri- posium I shall restrict myself to a consideroan and French Universities; Grants. from

ation of the more general aspects of the theory the Bache Fund....

481 and its simplest applications. For the sake

of logical completeness I shall have to refer to Scientific Notes and News...


many matters well known. The extension of University and Educational News...

484 the theory and its application to more special Discussion and Correspondence:

and complex cases, in so far as they can be Effect of Dormant Lime Sulfur upon the

handled on

his occasion, will be treated by Control of Apple Blotch: DR. E. F. GUBA. my colleagues. Crows and Starlings: F. R. WELSH. The

The first electrical theory of ferromagnetism Synchronal Flashing of Fireflies: OTTO was proposed by Ampère just about one hunA. REINKING. Franz Steindachner: DR. H.

dred years ago. On the basis of his own exW. WILEY


periments on the behavior of electric circuits Scientific Books :

and magnets, and on the assumption, already Groth's Chemische Krystallographie: PRO

justified, that magnetism is a molecular and FESSOR EDWARD H. KRAUS...

486 not a molar phenomenon, he concluded that

the molecule of iron is the seat of a perSpecial Articles:

manent electrical whirl and thus essentially a The Change in the Fat of Peanut-fed Rab

permanent magnet with its axis perpendicular bits: S. T. DOWELL,.

to the whirl. When the iron is fully magThe American Society of Mammalogists.... 487 netized, all the whirls are oriented alike, and

1 A paper read as a part of the symposium on recent progress in magnetism held at the joint meet

[ocr errors]

ing of the American Association for the AdvanceM88. intended for publication and books, etc., intended for

ment of Science, Section B, and the American review should be sont to The Editor of Science, Garrison-08- Physical Society, December, 1920. Revised, JanHudson, N. Y.



the magnetic moment of the mass of iron is moment of the whole molecule is zero in a the sum of the moments of the elementary neutral field. molecular magnets. Ampère undoubtedly con- In this case, which we shall consider in sidered that in a neutral mass of iron the some detail, the complete molecule will suffer molecular magnets are turned indiscrimi- no change of orientation when introduced into nately in all directions, but he did not enter a magnetic field, but the speed of the elecinto any discussion of the process by which tricity in each orbit or groove will change on their axes

are made parallel by the field account of the electromotive force around the during magnetization, nor did he consider the orbit or groove due to the alteration of the nature of the electrical whirls themselves. extraneous magnetic flux through it. Its Ampère was the grandfather of the electron magnetic moment д M

will thus increase (algetheory of magnetism. Wilhelm Weber was braically) by an amount All, which can readily its father. In 1852 Weber2 published a paper be calculated. If e denotes the charge of in which he developed a theory which, slightly electricity circulating in an orbit (whether as modified by Langevin,3 is still perhaps the a single electron, or a ring of electrons, or a most widely accepted theory of diamagnetism, continuous ring), m the mass associated with together with a theory of ferromagnetism the moving charge, r the radius of the orbit, which formed the starting point for the well- H the intensity of the extraneous magnetic known theory of Ewing. Weber adopted the field, and @ the angle between the axis of the molecular «whirls of Ampère, but assumed in orbit and the direction of the field, addition that these whirls, always present in

cos O.

(1) the molecules of magnetic substances, are also

4m present in the molecules of diamagnetic sub- If we assume that there are N orbits per stances when placed in a magnetic field. unit volume, all alike; and if we furthermore Further, he took the very important step of assume that all the orbits are perpendicular to attributing mass or inertia to the electricity the direction of the field (as they would be in in the whirls, and he assumed that the elec- the case of a saturated ferromagnetic subtricity moves as if in fixed circular grooves stance) we get for the magnetic moment of in the molecule, so that each whirl main- unit volume, or the intensity of magnetizatains its diameter and its orientation with tion: respect to the rest of the molecule as if rigidly


I = constrained. According to Weber's concep


4m tion, a substance is paramagnetic or ferro- and for the susceptibility magnetic when the molecule, or magnetic

e?r?N K = I/H

(3) element, contains a permanent whirl, with a

4m definite magnetic moment, and so tends to set If the orbits are not all perpendicular to the with its axis in the direction of any magnetic field intensity, but are uniformly distributed field in which it is placed; and a substance between all values of 6 from 0 to 7, as in an is diamagnetic when the molecule contains one isotropic diamagnetic substance, we get inor more frictionless grooves, with the mobile stead of (3) the expression electricity at rest before the creation of the

er?N K

(4) magnetic field. Langevin merely substitutes

12m electrons moving in fixed orbits for Weber's If in this equation we substitute the value electricity in grooves; and assumes that in a

of e/m known for electrons in slow motion, diamagnetic substance more than one orbit and assume for a given substance such as exists in the molecule and that the orbits are bismuth values of N and r which appear to be 80 constituted and grouped that the magnetic reasonable from other physical evidence, we 2 W. Weber's Werke, III., p. 555.

obtain from (4) values of K of the same order 8 Ann. chim. Phys. (8), 5, 1905, p. 70.

of magnitude as those found by experiment,

[ocr errors]

but the agreement is in general far from Thomson,4 Voigt,“ Lorentz, and others, inclose. The equation requires that K should cluding very recently H. A. Wilson.? If a be independent of the temperature, unless substance contains electrons either at rest or e, m, r and N depend upon it. As is well in plain rectilinear motion due to thermal known, the susceptibilities of many diamag- agitation, and a magnetic field is created, an netic substances are independent of the tem- electrical intensity will evidently be developed perature over wide ranges, while in other with a curl equal to the negative rate of incases there is a marked dependence.

crease of the flux density, which will cause According to this theory also, effects of the the electrons to move in paths curved in such same kind must exist in bodies which are a way as to produce a magnetic moment opferromagnetic or paramagnetic superposed on posed to the direction of the applied field; effects of opposite sign, the resultant sus- and as the field becomes steady curvature will ceptibility being, as Larmor long ago pointed be maintained by the action of the field on out, the sum of the two. The paramagnetic the moving electrons normal to their velociterm may account for the variation of the ties. Calculation on this hypothesis gives resultant susceptibility with temperature in susceptibilities of the same order of magnimany diamagnetic bodies. From Weber's tude as those given by the Weber-Langevin equation it may be shown that when A=0 theory. This form of theory has the advan


tages over the other of greater freedom from H,

(5) m4a

assumptions and of giving, when applied to where T is the period of the orbit. If we as

the optical case, a Zeeman effect with sharp

lines. Weber does not attempt to justify his sume that this period is that of sodium light,

assumption that in a molecule the diameters about 2 X 10-15 and that H=105 (in excess of any intensity hitherto produced) (5) gives

of his orbital grooves remain constant, and

that in a diamagnetic substance the grooves Au 0.3 x 10-3,

(6) maintain their orientations independent of the

applied magnetic intensity. With respect to 50 that the maximum diamagnetic effect is a

the diameters, however, Langevin has shown very small part of the saturation effect in

that the magnetic field will produce no alterferromagnetic substances. The fact that the

ation provided the law of force is not preintensity of magnetization of iron at satura

cisely that of the inverse cube, which is quite tion does not decrease appreciably even for

improbable. great increases of intensity shows that n=1/T

We shall return to the subject of diamagis very great.

netism later. From Weber's equations we may also calcu

The first detailed theory of paramagnetism. late the change in frequency n of an orbit due

was given for perfect gases by Langevin in to the magnetic field, and we find, after

1905.3 Following Langevin, I shall begin Langevin, but more generally,

with a gravitational analogue. Let us coneH cos 0

sider an enclosure containing a gas at uniAn

(7) 4лт

form temperature and let us suppose the This may correspond in a way to the Zeeman gravitational field anulled. The density of effect in light, but gives a broad band instead the gas will then be uniform throughout the of the sharp lines actually found, inasmuch as enclosure. If now the uniform gravitational cos A has all values between – 1 and + 1. field is brought into action every particle of

It is unnecessary, however, to have recourse gas will receive an acceleration downward, to electrons moving in orbits (or initially at

4 Int. cong. phys., 1900, vol. 3, p. 138. rest and constrained to grooves) or to rotating 3 Ann. der Phys. (4), 9, 1902, p. 130. electrified bodies, to explain the occurrence of B“The Theory of Electrons,diamagnetism, as has been shown by J. J. 7 Roy. Soc. Proc. A, 97, 1920, p. 321.

[ocr errors]

P. 124.

« iepriekšējāTurpināt »