November i6, 191 i] 



NATURE 



97 



SOCIETIES AND ACADEMIES. 

 London. 

 Royal Society, November 9.— Sir Archibald Geikie 

 I.C.B., president, in the chair.— Sir William Crookes : 

 'he spectrum of boron. The ph3'sical properties of the 

 'lament boron are almost unknown, notwithstanding the 

 fforts of many chemists who have worked on the subject. 

 >Ioissan, who came nearest to obtaining the pure element, 

 inly succeeded in getting it in the form of an amorphous 

 kowder. He said it was not possible to melt or volatilise 

 ,t in a carbon crucible or arc as it was changed into carbon 

 ^oride, and concluded that boron passed from the solid to 

 jhe gaseous state without becoming liquid. Recently Dr. 

 ,Veintraub, of the General Electric Company, U.S.A., has 

 \ot only obtained boron in a state of purity, but has pre- 

 pared it in a fused homogeneous state. His process 

 lOnsists in running an alternating-current arc between 

 Vater-cooled copper electrodes in a mixture of boron 

 ^hloride vapour with a large excess of hydrogen. The 

 ioron agglomerates on the ends of the electrodes, where 

 t grows in the form of small rods. After a while the arc 

 ^ns between two boron electrodes ; and if the current is of 

 ,roper value the rods melt down to boron beads, which 

 ventually fall off, whereupon the same process repeats 

 itself. The first specimens received from Dr. Weintraub 

 vere deposited from a vaporous state from boron chloride 

 Ind hydrogen in the manner described. Subsequently he 

 cindly sent the author some lumps of fused boron which 

 lad been prepared from magnesium boride. This boride 

 lissociates at a relatively low temperature (1200°), especi- 

 Uly in vacuo, and with rapidity at 1500°. The fusion is 

 •fTected between copper electrodes, the affinity of copper 

 or boron being so slight that it can be directly fused on 

 o the electrode without being contaminated with copper, 

 another way of fusing boron is in what Dr. Weintraub 

 ,alls a mercury arc furnace, based on the fact that most 

 jefractory bodies, such as tungsten, tantalum, boron, &c., 

 ave no affinity whatever for mercury. The result of the 

 uthor's work on boron is to show its photographed spec- 

 rum consists essentially of three lines, the wave-lengths of 

 vhich, according to accurate measurements, are 3451-50, 

 497-83. 'ind 2496-89. For more easy comparison the wave- 

 engths of these lines measured by' different observers are 

 ivpn below in a tabular form : — 



liariley (18S3) 



Rowland (1893) 

 I'-der and Valenta (1893) -■ 

 Exner and Haschek (1897) . 

 (1902).. 



34503 24970 2496-2 



— 2497-821 2496867 



345 1 "3 2497-7 24968 



345 1 '4 2497-8 2496-88 



3451-49 2497-79 2496-87 



Hagenbach and Konen (1908) 3451 2498 2497 



Ctookes(i9ii) ... ... 3451-50 249783 249689 



he fourteen other lines given by Eder and Valenta, and 



le five other lines given by Exner and Haschek, failed to 



'cord themselves on the photographs, notwithstanding 



^Mvely long exposures given in the attempt to bring 



idditional boron lines. The most interesting property 



. Mjlid boron is its extraordinary rise in electric con- 



uctivity with a slight increase in temperature. A piece 



f melted boron measured by Dr. Weintraub, which at 



le room temperature (27°) had a resistance of 5,620,000 



hms, dropped to 5 ohms at a dull red heat. Another 



■otevvorthy property of melted boron is extreme hardness. 



,t comes next to the diamond in hardness, a splinter easily 



- ^hing corundum. Its fracture is conchoidal, and no 



d crystalline structure is seen under the microscope. 



agglomerated boron deposited in the arc from boron 



de and hydrogen is, on the contrary, highly crystal- 



Hon. R. J. strutt : A chemically active modifica- 



if nitrogen produced by the electric discharge: H. 



'xygen destroys active nitrogen, but does not combine 



It. Hydrogen has no action. (2) Active nitrogen, in 



ng with nitric oxide to form the peroxide, gives the 



i^rcenish-yellow flame with continuous spectrum which 



l)e obtained by stimulating oxides of nitrogen in other 



(3) The reaction just mentioned is used to deter- 



the percentage of active nitrogen pr. v, ni in ordinary 



-;i'n as it loa,;es the discharge. The r.sull found is 



,3out 2-5 per cent., much higher than was formerly sup- { 



,ised. (4) When dilute phosphorus vapour is introduced . 



ito glowing nitrogen it does not react at once. It is not (' 



NO. 2194, VOL. 88] 



until some time attcr the glow has completely disappeared 

 that the nitrogen gets into a state in which it can react 

 with phosphorus. (5) The glow has a large electrical con- 

 ductivity, comparable with that of a salted Bunsen flame. 

 The ions are liberated in the glow, not merely carried for- 

 ward from the original discharge. This ionisation is, as 

 a rule, not very greatly affected when the spectra of other 

 substances, such as metals or cyanogen, are developed by 

 the active nitrogen in the space between the testing elec- 

 trodes. (6) None of these spectra are visibly diminished 

 in intensity when large electromotive forces are applied to 

 remove the ions. (7) Ozone can in some cases develop 

 metallic spectra when mixed at comparatively low tempera- 

 tures with the metallic vapour. — Sir J. Dewar : Produc- 

 tion of solid oxygen by the evaporation of the liquid. — Sir 

 J. Dewar and Dr. H. O. Jones : The gaseous con- 

 densable compound, explosive at low temperatures, pro- 

 duced from carbon disulphide vapour by the action of the 

 silent electric discharge : H. — Dr. T. H. Havelock : 

 Optical dispersion : a comparison of the maxima of absorp- 

 tion and selective reflection for certain substances. This 

 paper contains a discussion of various wave-lengths 

 associated with each dominant region in a general type of 

 dispersion formula. It is shown how the maxima of 

 absorption and of selective reflection are, in general, 

 separated from each other and from the wave-length corre- 

 sponding to the natural vibrations in the molecule. 

 Formulae are obtained for some of these maxima in terms 

 of the constants of the dispersion formula, and are con- 

 firmed by comparison v/ith available experimental results. 

 To estimate the magnitude of the differences in question, a 

 numerical study is made of regions of selective absorption 

 and reflection for carbon disulphide, rock salt, and sodium 

 vapour ; in particular, for rock salt it appears that the 

 maximum of selective reflection in the infra-red is dis- 

 placed considerably from the maximum of absorption and 

 from the dominant wave-length of the dispersion formula. 

 — Dr. T. H. Havelock : The influence of the solvent on 

 the position of absorption bands in solutions. According to 

 Kundt's rule, the effect of the solvent is to displace the 

 absorption bands further to the longer wave-lengths the 

 greater the refractive or dispersive power of the solvent. 

 By using a suitable type of dispersion formula this rule is 

 given a definite theoretical expression, and various experi- 

 mental results are examined from this point of view. 

 Although effects are complicated, in general, by molecular 

 changes, it is possible to estimate in some cases how much 

 can be ascribed to the operation of Kundt's rule. — Prof. 

 F. G. Donnan and Dr. J. T. Barker : An experimental 

 investigation of Gibbs's thermodynamic theory of inter- 

 facial concentration in the case of an air-water interface. 

 The " surface " concentration of a dissolved substance in 

 excess over that in the bulk of the solution is given by 

 Gibbs's equation V = —dirfdfi, where r = excess of solute 

 per unit of interface, <t =interfacial tension, /i = chemical 

 potential of solute. Assuming the simple osmotic law of 

 van 't Hoff for the solution, the above equation can be 



written in the form r = - =A-, ^-. where c = bulk concentra- 



tion of solute. The authors have tested this equation by 

 measuring independently r, c, and dff/dc for the case 

 of an air-water interface. The substances examined were 

 pelargonic acid and saponin. The value of F was deter- 

 mined by finding the change in concentration of a given 

 volume of solution caused by bubbling through it a known 

 volume of air in the form of a known number of very 

 small air bubbles. Steady streaming of the liquid was 

 prevented by breaking up the column of liquid into a 

 number of eddy chambers. The extremely small changes 

 of concentfation thus produced in excessively dilute solu- 

 tions were measured by means of a dropping pipette, the 

 same apparatus being also employed to measure dff/c/i; 

 The values of the two members of Gibbs's equation were 

 found to be in fairly good agreement, considering the 

 difficulty of the experiments. In the case of aqueous solu- 

 tions of pelargonic acid of concentrations varying between 

 0008 and 0-0024 gram per 100 grams of water, Ihe average 

 value of r found experimentally was in round numbers 

 one ten-millionth of a gram per square centimetre of inter- 

 face at 15° C. In the case of saponin the values found 

 were somewhat greater. 



