Supplement to '"Nature" May 5, 1904 



III 



SUPPLEMENT TO "NATURE." 



LORD KELVIN ON OPTICAL AND 

 MOLECULAR DYNAMICS. 

 Baltimore Lectures on Molecular Dynamics and the 

 Wave Theorv of Light. Bv Lord Kelvin, O.M., 

 G.C.V.O., P.C, F.R.S. Pp. xxii + 694. (Cam- 

 bridge : University Press, 1904.) Price 155. net. 

 I\ the autumn of the year 1884 Lord Kelvin delivered 

 at Johns Hopkins L'niversity a course of lectures 

 ■" On Molecular Dynamics and the Wave Theory of 

 Light," mainly extempore, which, having very fortu- 

 nately been reported stenographically by Mr. A. S. 

 Hathawav — one of his band of auditors, the famous 

 ■" twentv-one coefficients " — were issued to the world 

 unrevised in a papyrograph volume at the end of the 

 same j'ear, and have since been known as the " Balti- 

 more Lectures." The report, being nearly verbatim, 

 showed how comparatively slight were the immediate 

 preparations that Lord Kelvin had made for some 

 portions of his task, and thus had the great advantage 

 of revealing the procedure and attitude of an investi- 

 gator of transcendent genius in face of regions of his 

 subject more or less new to him. 



One result, in fact, of his enthusiasm for the new 

 aspect of optical propagation revealed by the pheno- 

 mena of anomalous dispersion, and of the wealth of 

 mechanical illustration which, taking his audience 

 into collaboration, he provided for this hitherto rather 

 abstract subject, was to start a period of enlivened 

 interest, illustrated by memoirs by Lindemann and 

 ■others, abroad and in this country ; this has now 

 reaped a reward in fruitful comparison of the theory 

 with experimental data obtained over the enormous 

 range of six or seven octaves by Langley, Rubens, and 

 other pioneers. Irrespective of this phenomenon of 

 anomalous dispersion in the spectrum near a region of 

 absorption of the light. Lord Kelvin's own estimates 

 of molecular dimensions had already ruled out the 

 ■earlier attempts of Cauchy and his followers to base dis- 

 persion on mere loading of the aether by massive 

 molecules. On such a theoni" of inert molecular 

 masses the proportional dispersion per octave could 

 idepend only on the ratio of the intermolecular distance 

 to the wave-length, no other magnitudes coming into 

 consideration, and it must depend on the square of 

 this ratio ; thus the actual 1 per cent, dispersion for 

 glass would be explicable by about. 10 molecules per 

 wave-length, while with the real number, about lo'', 

 the circumstances would be practically the same as 

 for a uniformly distributed load, which would give no 

 •dispersion at all. The modern theory of dispersion 

 thus must rest on an investigation of the interaction 

 between the forced internal vibration of the molecules, 

 conditioned by their own proper periods, and the 

 periodic impressed vibration of the wave-motion which 

 produces it. So long as the internal dynamics of the 

 molecule remain unexplored, only general principles 

 can be applied, and it matters little to the argument 

 whether it is conducted in terms of a mechanical con- 

 ception of radiation or in terms of the electric theory ; 

 in either rase only the genera! frame into which the 

 NO. 1801, VOL. 70] 



facts are to be fitted can be supplied by theory. On 

 a mechanical view, mere loading may produce refrac- 

 tion but not dispersion ; so on the electric view, even if 

 there were no free periods in the ordinary sense, there 

 would remain an index of refraction equal to the square 

 root of the dielectric coefficient. 



In the preface to the present volume Lord Kelvin 

 states that he chose for his lectures the subject of the 

 wave-theory of light with the object of accentuating 

 its points of failure, thereby intending to stimulate 

 the activities of his audience towards extending further 

 " the floods of new knowledge splendidly enriching 

 the whole domain of physical science " that had flowed 

 from the theory. 



" We all felt that difliculties were to be faced 

 and not to be evaded ; were to be taken to heart 

 ■with the hope of solving them if possible ; but at 

 all events with the certain assurance that there is an 

 explanation of every difficulty, though we may never 

 succeed in finding it. It is in some measure satis- 

 factory to me, and I hope it will be satisfactory to all 

 of my Baltimore coefficients still alive in our world 

 of science, when this volume reaches their hands ; to 

 find in it dynamical explanations of every one of the 

 difficulties with which we were concerned from the 

 first to the last of our twenty lectures of 1884." 



The sentences quoted contain the key to much 

 (though far from all) of Lord Kelvin's mathematical 

 in\-estigation of the last twenty years. The result is 

 this magnificent volume of more than 700 pages, which 

 in its variet}" of contents and width of grasp forcibly 

 recalls the original " Thomson and Tait " of forty 

 years ago, except, indeed, that the form of a treatise 

 being now dispensed with, the author is at liberty to 

 go directly for the various subjects of interest that 

 hold his attention without an\' necessity for pre- 

 liminary didactic expositions. 



The earlier lectures are reproduced here with addi- 

 tions within brackets, but the author soon found that it 

 was easier to re-write the greater part of the material. 

 His expression of distrust of " the so-called electro- 

 magnetic theory of light " (p. 45) stands as in the 

 original. Along with it is the interesting statement 

 that he had worked out for himself, as early as the 

 year 1854, the result that an electric impulse is pro- 

 pagated along a cable with a velocity of the order of that 

 of light, and that it only required a knowledge of the 

 ratio of the electric units to lead to the result that for 

 an air dielectric it would agree with the velocity of 

 light in air. An investigation of such linear propaga- 

 tion, on the lines now familiar, and thoroughlv de- 

 veloped by Heaviside, is inserted as the last appendix 

 (L). The first published determination of this velocitv 

 is contained, as is well known, in Kirchhoff's memoir 

 of 1857; there the result is deduced on the basis of 

 Weber's theory of moving electrons, which act on each 

 other instantaneously at a distance, the law of 

 attraction involving their velocities as well as their 

 distance apart. Neither there nor in the ordinary 

 modern investigation for cables is there anv reference 

 to transmission of electric effects across space with 

 finite speed; that makes no difference for the case of 

 enclosed cylindrical dielectrics of diameter small com- 

 pared with the wave-length, for with them it is only 

 the adjacent parts of the electric distribution and 



