,io 



NATURE 



[January 24, 1895 



determinations. Two arrangements of the kind described are 

 pUced side by side, and the same current is sent through their 

 two wires, which are joined in seiies and have approximately 

 equal resistances. The ratio of amounts of heat expended on 

 the two liquids divided by the ratio of the ma.<ses vaporised, is 

 equal to the ratio of their latent heats. The delerminalions 

 made by this method aj^ee well with those of other observers ; 

 but the authors' object is to obtain values correct wilhin about 

 per cent, for a large number of liquids, r.iiher than a highly 

 accurate value for any one substance. In reply to Mr. Griffith-', 

 the author; s'ated that the platinum wire was found to rise 

 about 20' above the temperaiuie of the liquid, and Mr. Griffiths 

 said that his experience had been similar. He did not see 

 why a very high degree of accuracy should not be obtainable 

 by the method. Prof. Kiicker expressed his admiration for the 

 work, and thought it justified by the fact that the results 

 accorded more nearly with theory than those of other o'servers. 

 — Mr. Eumorfopoulos read a paper on the determination of 

 thermal conduciviiy ami emissivity. In the fiist series of 

 experiments described, two bars of the same material and polish, 

 and each of uniform circular section, are heated, each at ofe 

 end, until the di.slribulion of temperature has become steady. 

 By means of two thermo-joints (one on each rod) a seiies of 

 isothermal points are then found. According to the ordinary 

 theory, if the two bars agree in temperature at a given pair of 

 points, they will also agree in temperature at distances .r, and 

 .r™ mrasurcd respectively from these pnints, where .Vj and .Vo 

 aie connected by the relation jr,/.r.j = ^'('■i/rol, r, and r.^ being 

 the radii of the rods. This relation was not lound to hold 

 good for the lods examined. In all cases .r,/ajwas further re- 

 moved from unity than the ordinary theory would lequire. One 

 conclusion was that the formula usually adopted in such ca'^es 

 could not be used for the comparison of conductivities, unless 

 the radii of the rods compared are equal, and their surfaces in I 

 the same condition. To settle the question, three lirass rods | 

 were chosen, and their absolute conductivities compared by j 

 Angstrom's method. The eniissivily was found to vary con- i 

 siderably with the radius, being greater the thinner the rod ; | 

 moreover, the value of the emisiivity deduced from the first 

 sine term of the Fourier's series was in each case found to be 

 about f2 limes as great as that deduced from the constant term. 

 — Mr. A. W. P.jrter then read a paper on the influence of the 

 dimensions of a body on the thermal emission from its surface. 

 The ordinary assumption is that whe'her a body is in varuo or 

 surrounded by air, the " emissiviiy " (/.<r. the amount of heat 

 pas-ing outwards from unit area per second per degree excess 

 of temperaiiire) is independent of the size of the body. Results 

 obtained experimentally by Peclet for cylinders and spheres of 

 difTcreiit sizrs, show that the emissiviiy depend- materially upon 

 the size of the body. Pcclei's formula; lor cylinders and spheres 

 surrounded by air show that for each of the c forms the r.ite of 

 emission per unit surface, excluive of ihe radiation efTeci, may 

 be represented by a constant plus a term inversely pr porlional 

 to the radium. The author examines the results of supposing 

 the lo45 to follow only in part the l.iw of radiation, the remainder 

 being assumed to follow the law of conduction. He thus 

 arrives at a formula 



a iiog K - log fl) 

 where e n the emissivity, a the radius of Ihe rod, K Ihe radius 

 of a hollow cylinder wlich surrounds Ihe bar, and above which 

 the excesses of lemptralurc arc rrcknncd ; while It and c arc 

 con>lant<. This f.irmuU has been compared with exiierimental 

 re«ull» of Ayilon and Kilgour, of M.icl-arlanc, of li utoniley, 

 and of PcJel, and has also lieen directly checked by experi- 

 ments on a brass rod when surrounded by wairr-jackets of 

 diflrrenl radii, as well as on the same bar unjickelrd. The 

 author finds the agreement lo be much closer than i* the case 

 on ihe ihrory of cmsiant emis-ivity, or with cmpi<ically de- 

 duced forirula nf Ayriori and Kilgour, and he concludes i hat 

 the enclo»ir>g boundary i- a» imp.irt.nni a facor in delermimng 

 emi"iviiy as the s\ie of the body inrlf. Pmf Carey Foster 

 ihouk'hi ibal in demonstrating the influence of the enclosure 

 Mr. Porter had e>lili||-hed an important point. Piof. Ayrton 

 .igreed a» lo the impoilance of the inflnrncc "f ih> enclosure. 

 lie urged ihal in »uih e«per>menl< as iho«e of Mr. Porter and 

 Mr. Kiimoifopiuios, the condMC'i>i'y and emissivity, which 

 were lumtions of the tcmprraiue, sliould not be asumcd 

 cooiiani along the bar. Mr. Tioller objected to the use of the 



NO. I317. VOL. 51"] 



term emissivity as including loss of heat by contact with the 

 air in addition to the loss by radiation. Mr. Griffiths said 

 that in some of his experiments, whrre a wire conveying an 

 electric current was immer-ed in a liquid in order to heat the 

 latter, the rise of the temperature of the wire above that of the 

 liquid was found to be ncaily independent of the diameter of 

 the wire. Mr. Kumorfopoulos said that in each case his com- 

 parison had been between portions of bars in which ihe range 

 of temperature was the same. Moreover, the variation of emis- 

 sivity and conductivity with lemperatuie, as found by other 

 ob-ervers, would be quite insufficient lo accoun' for his results. 

 Mr. Porter said that the term emissivity had come to be accepted 

 as referring to all heat lost at ihe surface of a body, whether by 

 radiation or by coniuclion and convection. In that sense he 

 had used the lerm. Prof. Rticker thoughl that emissivity, in 

 this sense, was not a good term, but to change now would 

 probably only m.ike greater confusion — Mr. G. U. Vule then 

 gave a brief outline ol his paper, on the passage of an oscillator 

 wave-tiain through a plate of conducting dielectric I3y a con- 

 ducting dielectric the author means a substance whose conduc- 

 tivity and dielectric capacity are both of importance in the case 

 under discussion ; and the paper is mainly an invest ii^ai ion of the 

 following problem : a rrain of plane elcclromngnetic waves falls at 

 normal incidence on an infiniie parallel sided plate of conduct- 

 ing dieleciric, whose thickness is finite, and at the first face of 

 th*- plate, the atnplitude of the vibration vect(»r in the incident 

 train is zero up lo a certain instant, aid then becomes equal to 

 an harmonic function of Ihe lime, mul iplied by an exponential 

 function with negative index: to find what pro| option of the 

 energy of the whole incident tr.iin is reflected luck, what pro- 

 portion is transmitted through the plate, and what i>roportion 

 absorbed. At successive inciHcnctS of reflected and rercflected 

 wave-trains upon the two bounding fnces of the plate, the am- 

 plitudes and phasechnngesof reflected and transmi'led portions 

 have to be taken into account, and the resnl'ing infiniie series 

 of terms have to be summed. The analysis is very long, but 

 the results obtained are exact. Curves are given, showing (for 

 spciial nunieric.il values of the constants of the problem) the 

 qua..i-periodic variation of the amounts of energy transmitted 

 and reflfced, as the thickness of the plate is increased from zero 

 up to a high v.Tlue. Oiher curves a'e given sho\* ing the eflfect 

 of varying the dii lectiic constant and the conductivity of the plate, 

 and the diflTcrence between a " damped" and an " undamped" 

 wave-train in regaid to intensity of refleced and transmitted 

 portions. The .luihor compares his cilculatcd results with 

 measurements obtained in the lase of oscillator waves travelling 

 along a dou'de-wiie cncuit about lOO metres in length; ihe 

 wires at the middle of the circuit being run ilirough a jar contain- 

 ing distilled water, alcohol, or a very ililiiie e'ectrolyte. — The 

 necessary corrections, hovever, are dfficuli and unccilain, and 

 Ihe author has not found it possible to deal with them in a satis- 

 factory way. — A letter from Dr. E. II. Harton was read, empha- 

 sising the necessity for taking into account the damping in the 

 oscillator-train, and at the same lime pointing out why, in his 

 opinion, the corrections applietl by Mr. ^^lle were inadetpiate 

 and failed to yield intelligilile results. Prol. Kiicker congratu- 

 lateit Mr. \'ule on his work, and on the iin|iortanceof ihe re-ulls 

 he had obiained. In returning the thanksoi ihe Society lo Prof. 

 Carey Foster for tlie invitation lo meet in Unveisi y College, 

 he expressed the pleasure he had fell in observiiig ihe extent 

 and compl tenes of the lalioraiories. Iliiheno London had 

 been hehiiid the provinces in this mailer, an I it was gratifying 

 lo finii that suulenis in London had now such opiior'unilies for 

 prac-ic.il instruction in physics. The papers wliith had been 

 reid at that m. eting were a proof that good u-e »as being made 

 of the lal'O'aioiics for the purposes ol research. The educa- 

 tional expeiimenis they had seen in the labora'ories were 

 excellently dtvised, and he hoped that many nf them would 

 become a part of the regular course of instruction in the 

 country. Prof. Fiser briefly replied. 



Chemical Society, Dec. 20, 1894, — Dr. II. K, Armstrong, 

 Presiil' n , in ilie chair. — The following papers were rend : — .^n 

 improved lomi of l-aroin-ter, bt J. N. Collie. The auttior has 

 devised a port^tble barouieler presenting several new ie;Uuies. — 

 The coni'iluenls ol I'ifer mjalum, by VV. K. Dunssin and H. 

 Gainelt, Piprr (n'alum is a VVcsl Indian medicinal plant ; 

 the au'hors have scpnraicd from it a lomc a kaliiidal subsiancc, 

 whiili lliey leim pipeiovaiine. C)(jl IjiNO.^, ; it seems likely lo 

 be of -eivicc in iheiapcuiics — Noie on iht- aciive cmsiituent of 

 the I'ellitory ol medicine, by VV. K, Dunsian and II, Gar- 



