Page images

adequate for the purpose of ascertaining extremely feeble inductioncurrents through diamagnetic substances. For this I combine with it a process of multiplication similar to Weber's, but with three


A commutator (I.) alternates in very rapid succession the direction of the current in A and B. A second commutator (II.), moved by the same mechanism, turns the conduction of the spirals a and b to the galvanometer in such wise that all the induction-actions of the diamagnet which arise in a on the alternation of the principal current, arrive in the same direction at the galvanometer. These induction-shocks (10-12 per second) give, according to known laws, a constant displacement from the position of rest. This enduring deflection can be directly observed. In harmony with all the investigations hitherto made, it appears that the arising and vanishing diamagnetism induces in neighbouring conductors currents opposite in direction to those obtained by means of magnetic bodies.

Observation becomes still more convenient and sensitive when another, a special commutator (III.) is brought into the galvanometer-conduction, which the observer, after the known method of multiplication, turns at the end of each oscillation until the amplitude reaches its extreme value. Of course small deviations from perfect compensation of the coils a and b must be determined by comparing observations with and without a diamagnet, and taken into account. In this way, with proportionally small spirals (A and B 500 turns each; a, b, and galvanometer each 1000), I obtained, the galvanometer being moderately astatized, a constant extreme amplitude of 15 scale-divisions by means of six Bunsen cells and inside the induction-spiral a bundle of rods of bismuth of 200 grms. weight; while a particle of fine iron wire weighing only 0.0044 grm. gave 556 scale-divisions in the opposite direction. With very delicate astatizing a single Bunsen cell was sufficient, with the above means, for the perception of the induction-current by the diamagnetism of the bismuth.

In this place I recommend also for similar observations a very simple modification of the mirror-galvanometer, by which four times instead of twice the angle of deflection is measured. Opposite to the galvanometer-mirror a fixed horizontal strip of lookingglass is placed, at a distance from 10 to 15 centims., and the telescope and scale adjusted so that the rays of light meet the galvanometer-mirror twice before entering the telescope-which is easily accomplished if the dimensions of the mirror are suitable. From the reading s the deflection-angle a is found according to the formula

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

in which D and d denote the distances of the scale and the fixed auxiliary mirror from the mirror of the galvanometer. This

arrangement is recommended in those cases where the measurement of the angle cannot be carried to a greater nicety by increasing the magnifying-power of the telescope or by large scale-distances.

In the above-described process for observing diamagnetism, beside the slight action to be measured, an incomparably greater one, the direct induction of the spirals, must be compensated. It is therefore a principal thing with the apparatus to be able to ensure the compensation and preserve it unaltered. This is only possible when the system of commutators (I. and II.) satisfies certain conditions in its construction, to be more particularly described in another place, and when the spirals are well insulated, so that in their windings no secondary closings variable with the temperature shall exist. Now these conditions, as we learn from observation, can be fulfilled with very remarkable completeness; and hence, in my institute, diamagnetic measurements will be carried out according to this method with more powerful spirals.

I have also executed another (and, so far as I know, undescribed) form of differential inductor, with which the induction produced by the arising and vanishing magnetism upon the principal current is observed. Let four branches A, B, C, D be, after the manner of Wheatstone's combination, united so that the principal current divides into two branches A+B and C+D, and let the bridge together with the galvanometer be inserted between the angle-points AB and CD. Let the branches A and C contain each a single spiral of great magnetizing force. Now let the action of the stationary current upon the galvanometer be compensated by resistances in B and D. The action, in general still present, of the extra currents at the closing and opening of the circuit is compensated separately, by micrometrically displacing fine iron rodlets in the less powerful spiral until the galvanometer shows neither stationary nor momentary deflections. Induction, however, is immediately again produced by the opening and closing when a feebly magnetic body is placed in the other spiral; this can, as before, be multiplied and measured.

It is true that this last method is far more difficult in practice; but it might be suitable for the investigation of certain reactions on current-electricity. Thus, for example, by it one might study the question whether the rotation of the polarization of light in magnetic or diamagnetic bodies under the influence of the current reacts on the current, which, with present views on electricity and the luminiferous æther, is not improbable. It was intended observations of this kind that induced me to make the above-described preliminary experiments on electrical induction-currents through diamagnetic bodies.-Kaiserliche Akademie der Wissenschaften in Wien, Sitzung der math.-nat. Classe, January 21, 1875.











APRIL 1875.

XXIX. On the Comparison of some Tube and other Spectra with the Spectrum of the Aurora. By J. RAND CAPRON, Esq., F.G.S.* [With Four Plates.]


[ocr errors]

Na contribution by the late Professor Ångström to the solution of the problem of the aurora-spectrum (an abstract of which appears in Nature' of the 16th July, 1874), the Professor is stated, amongst other things, to have laid down certain propositions in substance as follows:

1st. That the aurora has two different spectra-the one comprising the one bright line in the yellow-green only, and the other the remaining fainter lines.

2ndly. That the bright line falls within a group of hydrocarbon lines, but does not actually coincide with any prominent line of such group, and that Herr Vogel's finding this line to coincide with a not well-marked band in the air-spectrum must be regarded as a case of accidental coincidence.

3rdly. That moisture in the region of the aurora must be regarded as nil, and that oxygen and hydrogen must alone there act as conductors of electricity. Professor Angström then details the examination of an exhausted dry air-flask filled with a discharge analogous to the glow of the negative pole of a vacuum air-tube, and, comparing this with the aurora-spectrum, gives the following results:

Aurora-lines, wave-lengths 4286 4703 5226
4272 4707 5227

Violet light,

[ocr errors]

Two weak light bands found by Herr Vogel at 4663 and 4629

Communicated by the Author.

Phil. Mag. S. 4. Vol. 49. No. 325. April 1875.


« PreviousContinue »