344 



SCIENCE-GOSSIP. 



conducted by james quick. 



Peoduction of High Vacua by Liquid Hydro- 

 gen. — Professor Dewar's investigations upon 

 liquid hydrogen and the application of it are 

 opening up a wide field both to physicists and 

 chemists. Not the least interesting part of the 

 work is the rapidity with which by its means high 

 vacua canbeprodnced in vessels. Asamatter of fact 

 the vacua obtained are so perfect that it is inferred 

 both theoi'etically and experimentally that the 

 vacuum left after liquefying the air out of a 

 vessel by means of liquid hydrogen cannot exceed 

 the millionth part of the atmospheric pressure, 

 excluding the pressure from any incondensible 

 material other than nitrogen and oxygen. Two 

 vacuum tubes were taken and .arranged so that 

 their drawn-out open ends could be inserted in 

 liquid hydrogen and the air contained in them 

 solidified. They were then sealed off. On attempt- 

 ing to pass an electric spark through them their 

 excellent exhaustion was revealed by great resis- 

 tance to the passage of the discharge. Similar 

 tubes specially freed from gases and impurities on 

 the glass, had to be heated before discharge would 

 take place at all. The rapidity with which the 

 exhaustion in these tubes takes place is very great, 

 only a few seconds required to make it complete. 

 If it becomes possible to perform this kind of work 

 on a larger scale, and in a less expensive manner, 

 the extensive manufacture of vacuum tubes for 

 Eontgen Pay and other work will be very materi- 

 ally simplified and shortened. 



The Neenst Electeic Light. — In a paper 

 before the Society of Arts on February 8th 

 last Mr. J. Swinbiu-ne gave a detailed descrip- 

 tion of the electric lamp lighting, recently 

 invented by Professor Nernst, of Gottingen. The 

 present system of incandescent electric lighting- 

 is the heating of a thin carbon filament of different 

 thicknesses and shapes to conform to the various 

 requirements of pressure, candle-power, &c. The 

 question of incandescent lamps is one that has not 

 received the attention it merits. Since this form 

 of lighting began to be extensively used, not much 

 has been done to radically improve the efficiency. 

 Incandescent lamps of the present day still require 

 from 2^ to 4 watts per candle-power for satisfac- 

 tory lighting. Nernst rejects carbon altogether 

 in his form, and uses highly refractory oxides as 

 his material, as is done in the Welsbach lights. 

 The substance is an insulator at ordinary tempera- 

 tures ; but becomes an electrolyte at high tem- 

 peratures. Upon this fact the working of the 

 light depends. The oxides are made up in the 

 form of thin rods, having two platinum wires con- 

 nected to the ends. As the material is at first an 

 insulator, it requires to be heated to a compara- 

 tively small extent by some external means before 

 the current can pass and raise it to white incan- 

 descence. The arrangement takes the form of a 

 heating resistance — close to the rod and in 



connection with it. This circuit is broken 

 again as soon as the rod is heated suffi- 

 ciently to conduct, when the main current 

 then raises the temperature of the rod, and there- 

 fore the light emitted, to far beyond that attained 

 with the ordinary carbon filament. The efficiency 

 is remarkable, the average consumption being 

 I'o watts per candle power. If no serious difficul- 

 ties occur in the commercial working of this prin- 

 ciple, the importance of it to the electric lighting- 

 industry cannot be overrated. 



Fawcett's Patent Standard High Eesistances. 

 — The two forms of high resistances in use at 

 present are those made ■^\'ith insulated wire and 

 those with carbon lines upon an insulating base. 

 The first of these is costly, the second unreliable. 

 Mr. F. B. Fawcett has done a considerable amount 

 of work upon this subject and has now brought 

 forward an improved method for the making of 

 these resistances, and one which has satisfactorily 

 stoodsevere tests. The resistance films are metallic 

 and are produced upon glass by the electrical dis- 

 charge of the metal particles from a cathode com- 

 posed of a grid of platinum and geld ; the whole 

 arranged in a vacuum. When the film is first 

 deposited, its resistance alters at a rapidly in- 

 creasii] g rate ; the alteration continuing f ( r many 

 months. This is probably due to dissolved gas 

 and molecular rearrangement in the film. The 

 resistance becomes perfectly constant, however, 

 if the film is boiled for several hoiu'S under 

 diminished pressure in a suitable oil. Experiments 

 were made by Mr. Fawcett to arrive at the effect, 

 upon the temperatiu-e co-efficient, of varying the 

 thickness of film used. Starting with a film the 

 thickness of which we may, comparatively, call 1, 

 the alteration in resistance per degi'ee C. rise in 

 temperature was "0028 ; with a film of thickness 

 99 the value was -0153, thus showing the gi-eat 

 advantage of thin films. By diminishing the thick- 

 ness still further the temperature co-efficient has 

 been decreased to '0004 per cent., which is prac- 

 tically 7iil. The exact resistance of any film is ob- 

 tained by a process of scratching with a needle 

 when the film has been hardened by its immer- 

 sion into the oil bath. Further advantages of 

 these resistances, besides those given above, are 

 their negligeable capacity and self-induction. 



Measuring Extreme Temperatures. — This 

 subject formed the discoiu'se by Professor 

 Callendar at the Friday evening meeting of the 

 Eoyal Institution on March 10th. The measure- 

 ment of high temperatures has been and is an 

 absorbing subject with Professor Callendar, to 

 which the number of instruments specially 

 designed by him, both on the lecture table and 

 exhibited in the library, bore sti'iking e-^ddence. 

 The difficulties in the way of deducing high tem- 

 peratures not obtainable by direct experiment, 

 were pointed out. "Extrapolation" as a means 

 of obtaining these temperatvu-es has led to con- 

 fusing results and curves, due to insufficient data ; 

 and Professor Callender remarked that the only 

 way out of the difficulty was to extend the range 

 of actual observation and measiu-ement. Many 

 experiments were shown to illustrate the 

 different points, amongst them being an arrange- 

 ment for sho-wing, -with, gases, the principle of the 

 Wheatstone bridge. Illustrating this principle 

 by the flow of water is not new, but it is interest- 

 ing to see an extension of the application. 



