PHYSICS, PROGRESS OF, IN 1893. 



625 



gated (ho magnetism of iron rings slit radially. 

 'I'll.' -lit \v;i- u-i IM :!-57 millimetres wide, and 

 the strength of the nwLrnetie field l to yoo. Ho 

 finds i hi- demagnetizing factor constant up to 

 about ono half saturation, and the coefficient of 

 dis|x-rsion that is, the ratio of the mean induc- 

 tion to that at the slit constant to the' same 

 point. The coefficient increased with the width 

 of the slit, and finally decreased with inert-used 

 intensity of the field, but was Independent of 

 the radius of the ring. The region of dispersion 

 of force lines was limited essentially to the vicin- 

 ity of the slit, and was narrower as the magnet- 

 ization increased. G. M. Minchin (" Philosoph- 

 ical Magazine," August), in treating of the 

 magnetic field of the electric current, states 

 that its magnetizing effect is not the same at 

 points near the surface of the conducting win- 

 as if the whole current were concentrated in a 

 thin filament at the center of the wire. Such 

 an assumption -is allowable only when the wire 

 is straight or when it makes a curve of very 

 large radius. Prof. Ewing and Miss Klaasen 

 (London Royal Society, June 1), in experiments 

 with Ewinjj's magnetic curve-tracer ("Annual 

 Cyclopedia, 1 ' 1892, p. 642), find among other re- 

 sults that the time lag in soft, thick bars is im- 

 mensely great. The work spent per cycle is a 

 maximum at a particular frequency, which in 

 such bars is very low. Heydweiller (" Sitzung- 

 berichte of the Wlirzburg Physical Society," 

 March 11) proves the existence of a Villari point 

 in nickel hitherto unknown. The magnetization 

 used was very feeble, and the observing device 

 very sensitive. The magnetization under 2 C. 

 G. S. units shows a decrease with the smallest 

 loads and an increase with the larger. The 

 point seems to be between 765 and 246 grammes 

 of load for I = 0'97. 



Magnetism of Oxygen. Dewar, in a Royal 

 Institution lecture ("Nature," May 25), placed 

 liquid oxygen in a cup of rock salt and drew it 

 up with a magnet till the poles were connected 

 by the liquid. Liquid air acted in like manner, 

 but the oxygen and nitrogen were not separated 

 by the action of the magnet. P. Curie (Societe 

 Francaise de Physique, April 21) finds that the 

 magnetic permeability of oxygen is constant for 

 forces varying between 200 and 1,350 units and 

 pressures of 5 to 20 atmospheres. Between 20 

 and 450 it varies inversely as the absolute tem- 

 perature. 



Chemical Action in a Magnetic Field. Lieut. 

 G. O.Squier, U.S. A. ("Philosophical Magazine." 

 June), finds that when iron is exposed to chemical 

 action in a magnetic field there are two opposite 

 effects : (1) A protective effect due to the direct 

 influence of magnetism on the metal, and (2) the 

 concentration of products of the reaction about 

 the more strongly magnetized parts of the iron, 

 tending to produce higher potential, and finally 

 establishing permanent electric currents. 



Diamagnetism. P. Curie (Paris Academy of 

 Science, Jan. 23) has measured the permeability 

 of a series of diamagnetic bodies by inclosing 

 them in an exhausted glass vessel exposed to a 

 magnetic field and then repeating his experi- 

 ment with the glass alone. Most of the bodies 

 examined showed a constant coefficient, water 

 and 'quartz not varying perceptibly with the 

 temperature. That of potassium nitrate was the 

 VOL. xxxni. 40 A 



same for the solid and the liquid form, but that 

 of bismntli fell till the point of fusion (ti i 

 was readied, when it suddenly dropped from 

 !ir>7 to -IMS, and then remained consUml. 1'rof. 

 Fitzgerald (London Physical Society, Oct. 28) 

 says that diamagnetism corresponds to electro- 

 siatif induction, but that paramagnetism has no 

 tit-finite electric analogue, and is probably con- 

 nected with an arrangement of material mole- 

 eules, while diamagnetism depends on their 

 electric charges. 



Magneto-rotatory Phenomena. Kundt (Ber- 

 lin Physical Society, Feb. 10) has observed the 

 change of magnetic rotatory power in metallic 

 films with change of temperature. With nickel 

 there was no change at first with rise of tem- 

 perature, but above 300 there was a sudden 

 diminution, which became greater as the tem- 

 perature continued to rise, its relation to increase 

 of temperature being the same as for the mag- 

 netic susceptibility of the metal. 



Residual Magnetism. Ewing (" Philosophical 

 Magazine," October, 1692), in experiments on 

 the influence of joints in reducing residual mag- 

 netism in iron, finds that the division of a ring 

 80 centimetres long into two half rings abutting 

 against each other, with smooth joints, reduces 

 it from 9,000 to 6,000. 



Magnetic Viscosity. Hopkinson, Wilson, and 

 Lydall (London Royal Society, April 20) find 

 that after a sudden change of the magnetizing 

 force the induction sometimes does not attain its 

 full value for several seconds, and that the differ- 

 ence between a ballistic curve of magnetization 

 with complete cycles and a curve with consider- 

 able frequency is due to a true time effect. 



Miscellany. Magnetic and Electric Instru- 

 ments. G. Quincke (Wiedemann's "Annalen," 

 No. 1) has devised a novel form of instrument, 

 which may be used either as a tangent galva- 

 nometer or as a magnetometer. A glass disk is 

 supported vertically, and around the rim wires 

 are clamped. In a hole in the center is a mirror 

 with a magnet. This device has the advantage 

 of great simplicity. 



Dimensional Units. Ostwald (" Journal of the 

 Chemical Society," October, 1892) suggests the 

 use of energy as a dimensional unit. The result 

 would be in the direction of simplification, as the 

 following table shows : 



Reduction of Dimensions. J. Del- 

 boeuf (Belgian Royal Academy, No. 6) discuss, s 

 what would happen if the entire universe were 

 reduced proportionately in dimensions. It has 

 been urged that such a process would not be per- 

 ceptible to us, since all our means of estimating 

 size are only relative, and hence, for aught we 

 know or ever can know, the universe may be ex- 



S uiding or contracting at any rate of speed, 

 ut M. Deltweuf shows that we would have 

 several ways *of detecting such a change. For 

 instance, if the universe should shrink to half its 



