March 20, 1902] 



NA TURE 



47; 



form of the original Fizeau apparatus. The observing part of 

 the instrument is separated from the expansion chamber, and 

 the temperature of the interference tripod and the substance 

 under investigation, which it carries, is measured by means of 

 a thermometer bent just above the cylindrical bulb and so 

 arranged that the latter lies on the tripod table. The chief 

 advantages over the Fizeau apparatus are briefly, (l) the em- 

 ployment of a micrometric method of measuring the position 

 and width of the interference bands ; (2) the use of autocolli- 

 mation ; (3) the employment of C hydrogen light ; (4) an 

 arrangement of the thermal chamber which readily permits an 

 extension of the range to 120 C. The author has also intro- 

 duced an aluminium compensator, a relatively thick disc of 

 aluminium laid on the top of the porcelain tube. This over- 

 comes the difficulty of polishing the porcelain and affords a 

 large field of bands instead of an annular ring showing parts of 

 bands. The mean of three determinations with three specimens 

 of material gives the following result for the linear expansion, 

 L, = L„[i + io~*(2522/ -f- 7"43/")]. The results presented by 

 the author agree tolerably well with those of Chappuis, but the 

 constant a is slightly smaller. The constant b is seven times 

 larger than according to Bedford. The discrepancies between 

 the results of Chappuis and Bedford appear to be due to a 

 fundamental real diflerence, dependent on the interval of tem- 

 perature for which the determinations were made. The incre- 

 ment per degree of the coefficient of expansion of porcelain is 

 not a constant quantity, but one which is much larger 

 between o' and 100" than at higher temperatures. 

 The supposition of Prof. Callendar, as to the anomalous expan- 

 sion between 0° and 100°, appears to be well founded. — -The 

 secretary then read a paper by Mr. W. Williams on the temper- 

 ature variation of the electrical resistances of pure metals and 

 allied matters. In the first part of the paper an attempt is made 

 to correlate the periodic variations which pure metals exhibit as 

 regards their atomic weights, chemical valencies, melting points 

 and electric resistances. If m is the chemical valency, V the 

 atomic volume, 9 the absolute temperature, T the absolute melt- 

 ing point and c, or V^aT, the constant of Pictet's law, then 



a 00 , where <r is the specific resistance at 0° C. This 



<rT ^ 



relation holds for most of the metals, but fails for gold, 

 indium, tin and aluminium, and also for metals of the 

 iron group. The temperature resistance-coefficients of pure 

 metals are not equal to 1/273, ^^^ ^n expression for 

 the change of resistance with temperature has been de- 

 duced which holds approximately for many metals. The 

 author also obtains simple expressions for the average increment 

 per degree of the specific heat of metals and for the ratio 

 between the specific resistances of the solid and liquid states of 

 a metal at the temperature of fusion. — A paper entitled "A 

 suspected case of electrical resonance of minute metal particles 

 for light waves ; a new type of absorption," by Prof. R. W. 

 Wood, was read byi the secretary. Experiments on which the 

 author has been engaged have led him to believe that he has 

 found a new type of light absorption, which it may be possible 

 to refer to the electrical resonance of small metallic particles 

 for waves of light. Metallic deposits on glass have been pro- 

 duced which are shown by the microscope to consist of particles 

 less than the wave-length of light, and which by transmitted 

 light exhibit colours as brilliant as those produced by aniline 

 dyes. The author has sought to explain these colours by inter- 

 ference and diffraction, and has been forced to accept the 

 hypothesis suggested in the title of the paper. The metallic 

 deposits can be obtained by heating small fragments of the 

 alkali metals in exhausted glass bulbs, when the vapour con- 

 denses on the cold parts of the bulbs and forms the films. It 

 can be shown that the colours are due to the presence of 

 metallic sodium (in the case in which sodium has been used) by 

 allowing air to enter the bulb ; oxidation takes place and the 

 film vanishes. In some experiments the air has been allowed 

 to enter very slowly, and the changes which the film undergoes 

 before it vanishes have been examined. The particles which 

 form the deposits can be classed under three heads : (I) coarse 

 particles which diffract or scatter light and give the bulb a 

 silky lustre ; (2) minute particles very close together which 

 regularly reflect those wave-lengths absent in the trans- 

 mitted light, but give no scattered light ; and (3) 

 minute particles far apart which diffuse light of the 

 same wave-lengths as those which are to some extent 

 absent from the transmitted light. By observing the spec- 



NO. 1690, VOL. 65] 



trum of the transmitted light, the author has examined 

 the changes in colour which accompany changes in tempe- 

 rature of the film;. The paper gives an account of the 

 relation between the colour of the film and the size and distri- 

 bution of the particles, and also of the behaviour of the films 

 with polarised light. Experiments upon the electric resistance 

 of the films have proved that they are non-conducting. The 

 author concludes by stating that at the present stage it is im- 

 possible to decide either in favour of, or against, the theory of 

 resonance. The idea of resonance has proved a useful working 

 hypothesis for explaining some of the phenomena described in 

 the paper. The secretary read a letter from Prof. R. Threlfall 

 directing the attention of the author to some experiments upon 

 the same subject published by him in 1894. 



Chemical Society, March 6.— Dr. E. Divers, F.R.S., 

 vice-president, in the chair. — The slow oxidation of methane at 

 low temperatures, by Dr. Bone and Mr. R. V. Wheeler. Much 

 controversial matter has been published on this subject, the 

 point in dispute being which of the two constituents of marsh 

 gas, viz. carbon and hydrogen, undergoes oxidation first when 

 the gas is burned in an insufficient supply of oxygen. The 

 authors find that methane between 300' and 400° C. burns simul- 

 taneously to carbon monoxide and water, small quantities of 

 carbon dioxide being also formed, probably as the result of a- 

 secondary reaction. — Isomeric additive compounds of dibenzyl 

 ketone and deoxybenzain with benzal/-toluidine, «-nitrobenzil- 

 aniline and benzal-w-nitraniline, iii., by Dr. F. E. Francis. — 

 These additive compounds exist in three modifications, differing in . 

 melting point and other physical properties, and are thus easily 

 separable one from another. — Mesoxalic semi aldehyde, by Messrs. 

 Fenton, F.R S., and Ryffel. When tartaric acid is treated- 

 with gaseous chlorine in presence of ferrous salts it undergoe.'i 

 oxidation, forming the semi-aldehyde of mesoxalic acid, which 

 can be isolated in the form of its osazone and dioxime. — The 

 action of hydrogen peroxide on carbohydrates in presence of 

 ferrous salts, by Messrs. Morell and Crofts. In this reaction, 

 glucose, mannose, Isevulose and arabinose are oxidised to their 

 corresponding osones. — ;«-Nitrobenzoyl camphor, by Dr. M. O. 

 Forster and Miss F. M. G. Micklethwait. This substance has 

 been prepared by the action of nitric acid on benzoyl camphor. 

 The most interesting feature exhibited by it is that it only exists 

 in the enolic form, whilst benzoyl camphor occurs in both 

 ketonic and enolic modifications, camphor itself being only known 

 in the ketonic form. — The Cloez reaction, by Dr. Chattaway 

 and Mr. J. M. Wadmore. When cyanogen chloride or bromide 

 reacts with sodium ethoxide, urethane and triethylcyanurate are 

 formed. This reaction is, at first sight, inexplicable if the 

 formula C:N.C1 be assigned to cyanogen chloride, but the 

 authors suggest that the latter reacts first with the alcohol, form- 

 ing an ethyl iminochlorocarbonate, which in secondary changes 

 gives rise to the products mentioned above. — ^The picrimido- 

 thiocarbonic esters, by Mr. J- C. Crocker. The author describes 

 the products obtained when picryl chloride reacts with ammonium-. 

 thiocyanate. — Robinin, violoquercitrin, myrticolorin and osyri- 

 trin, by Mr. A. G. Perkin. The author finds that myrticolorin, a 

 quercetin glucoside, is identical with osyritrin. — The nitration 

 of .tywtrihalogen anilines, by Dr. Orton. When the aniline 

 operated upon contains a bromine atom in the para-positions 

 relative to the amino-group, it is replaced by a nitro-group. 

 The similar chlorine substituted anilines do not undergo this 

 replacement. — Some j-j/w.nitrochlorobromanilines and their 

 derivatives, by Dr. Orton. A description of the products 

 obtained in the course of the foregoing investigations. — The 

 resolution of pheno-a-aminoheptamethylene into its optical 

 merides, by Prof. F. S. Kipping, F.R.S.,and Mr. A. E. Hunter. 

 This base undergoes de-racemisation when it is converted into 

 the racemic (/-tartrate and the latter is fractionally crystallised 

 from water containing tartaric acid. 



Linnean Society, February 20. — The Rev. T. R. R, 

 Stebbing, F.R.S., in the chair.— On behalf of Mr. G. M. 

 Thomson, of Dunedin, N.Z. , the secretary exhibited a series of 

 photographs of New Zealand flowers. In connection with the 

 plants, some observations were made on the birds which visit 

 them, e.g. the bell-bird or "korimako," Anthornis inelaniira, 

 the grey warbler, Gerygoiic flavirostris, the pied fantail, 

 Rhipidura flabellifera, and the yellow-breasted tit, Pelroeca 

 viacrocephala. Of these, the first-named was observed to assist 

 in the fertilisation of the native fuchsias, on quitting which the 

 feathers of the head were seen to be stained with the bright 



