Oct. 25, 1883] 



NATURE 



625 



tensity of light during the next period of ab ence of moonlight 

 is slightly less, but we continue an ephemeri. from the elliptic 

 elements calculated by M. Fabritius : — 



At Gre.nwich Midnight 



The Variable Star U Cephel— A minimum of this 

 variable was observed by Mr. Knott at Cuckfield, on the 

 evening of October 20. The time was 8h. 34m. G.M.T., and 

 the star's magnitude was 92. The minimum fell an hour later 

 than Schmidt's elements (A. N. 2382) would predict. The 

 divergence of Mr. Knott's observations has increased to that 

 amount from nine minutes in 1881 ; at the same time he doubts 

 if a light increase of the adopted period would of itself c >m- 

 pletely saiisfy the observations, and perhaps the period may be 

 subject to variation. 



Reckoning from the minimum on October 20, and using 

 Schmidt's mean period, the next few minima will fall thus : — 

 October 30, 7I1. 52m. G.M.T. ... November 9, 7I1. urn. 

 November 4, 7b. 31m. G.M.T. ... November 14, 6h. 50m. 



PHYSICAL NOTES 



At the British Association meeting a paper by Prof. J. A. 

 Ewing was read, on the magnetic susceptibility and retentivcness 

 of iron an 1 «teel. This paper was a preliminary notice of some 

 results of an extended investigation which the author had been 

 conducting for three years in Japan. Experiments with annealed 

 rods and rings of soft iron wire showed that that material 

 pos esses the property of retentiveness in a very high degree. As 

 much as 90 and even 93 per cent, of the induced magnetism survived 

 the removal of the magnetising force. The extiaurdinary spec- 

 tacle was presented of pieces of soft iron entirely free from 

 magnetic influence nevertheless holding an amount of mag- 

 netism (per unit of volume) greatly exceeding what is ever held 

 by permanent magnets of the best tempeied steel. The mag- 

 netic character of the iron in this condition was, however, highly 

 unstable. The application of a reverse magnetising force quickly 

 caus;d demagnetisation, and the slightest mechanical disturb- 

 ance had a similar effect. Gentle tapping removtd the residual 

 magnetism completely. Variations of temperature reduced it 

 greatly, and so did any application of stress. On the other hai.d, 

 the magnetism disappeared on y very slowly, if at all, with the 

 mere lapse of time. The residual magnetism in hardened iron and 

 steel was much less than in soft annealed iron. The maximum 

 ra:io of intensity of magnetism to magnetising force during the 

 magneti-ation of soft iron was generally 200 or 300, and could 

 be raised to the enormous figure of 1590 by tapping the iron 

 while the magnetising force was beirg gradually applied. A 

 number of absolute measurements were made of the energy ex- 

 pended in carrying iron and steel through cyclic change- of mag- 

 net. sation ; and the effects of stress on magnetic susceptibility 

 and on existing magnetism were examined at great length. The 

 whole subject was much complicated by the presence of the 

 action which, in previous papers, the writer had named Hysttre- 

 sis, the study of which, in reference both to magneiism and to 

 thermoelectric quality, had formed a large part of his work. 



M. 1'. TlHON has lately shown at the Industrial Science 

 Society of Lyons a new semi-incandescent lamp, giving the 

 brilliancy of an arc light. This is attained hy having two carbon 

 rods, slightly inclined to one another, brought down on to a 

 small prism of chalk, and separated from one another by a small 

 rod of the same material. The current passes through the chalk 

 rod making it incandescent. By this means the light is rendered 

 steadier than an arc light, and it is said to have the same 

 brilliancy. 



Mr. Frank Gekaldy has published some interesting statis- 

 tics comparing the cost of the eltctric light with gas, both as to 

 its actual cost and its cost per candle power : — 



S 2 



§6 



t/3 



euo"H 



gss. 



= « 



« ' — 



6 5 



B : a 



« « 



x 



a." 



CO 



W 



o ? 



a 



s 



e 



w 





This is only an extract from a longer li.t, but conclusively shows 

 that in large instalment, electric lighting is cheaper than gas on 

 the to'al cost ; whilst considered per candle power it is far away 

 cheaper. An exception to the rule seems to occur in the first on 

 the list ; this is due to the smallness of the installation. In the 

 case of the Thames Embankment the light is reduced by the use 

 of ground glass globes. If we bear in mind the fact that the 

 economy consists in having large installations, we shall be 

 brought face to face with the fact that whereas gas is now made 

 in as large quantities as is practicable, electricity has still to be 

 brought to that state of economy. Thus we may still expect a 

 greater economical advantage than is shown by the above 

 figures. 



M. J. Jamin has a paper in the Journal de Physique on the 

 " Critical Point of Liquefiable Gases," in which he discusses ? 

 neiv theory. He says : " I believe that gases are liquefiable a 

 all temperatures when the pressure is sufficient." Describing 

 Cagniard-Latour's experiment, he says : "According to knowr. 

 laws, the quantity of vapour above the liquid increases very 

 rapidly, its density increasing at the same rate as its weight 

 without known limit. Again, the remaining portion of the 

 liquid expands at an increasing rate until it passes that of the 

 gas (Thilorier) ; it is clear then, by the effect of these inverse 

 variations, that at last a limiting temperature must be reached 

 when the liquid and the vapour must have the same weight for 

 the same volume. At this point they are inseparable; the vapour 



