424 



SCIENCE. 



[N. S. Vol. XII. No. 299. 



agation, the density of the radiant energy 

 in the medium at any place must be ex- 

 tremely small in comparison with the 

 amount of energy that is transmitted in a 

 second of time : this easily led him to the 

 very striking conclusion that, on the hy- 

 pothesis that the aether is like material 

 elastic media, it is not necessary to assume 

 its density to be more than 10"'* of that of 

 water, or its optical rigidity to be more 

 than ten 10"* of that of steel or glass. Thus 

 far the sether would be merely an impal- 

 pable material atmosphere for the transfer- 

 ence of energy by radiation, at extremely 

 small densities but with very great speed, 

 while ordinary matter would be the seat of 

 practically all this energy. But this way 

 of explaining the absence of sensible influ- 

 ence of the sether on the phenomena of ma- 

 terial dynamics lost much of its basis as 

 soon as it was recognized that the same 

 medium must be the receptacle of veiy high 

 densities of energy in the electric fields 

 around currents and magnets.* The other 

 mode of explanation is to consider the sether 

 to be of the very essence of all physical ac- 

 tions, and to correlate the absence of ob- 

 vious mechanical evidence of its interven- 

 tion with its regularity and universality. 



On this plan of making the sether the 

 essential factor is the transformation of 

 energy as well as its transmission across 

 space, the material atom must be some 



*We can here only allude to Lord Kelvin's recent 

 most interesting mechanical illustrations of a solid 

 sether interacting with material molecules and with 

 itself by attraction at a distance : unlike the general- 

 ized dynamical methods expounded in the text, 

 •which can leave the intimate structure of the material 

 molecule outside the problem, a definite working 

 constitution is there assigned to the molecular nu- 

 cleus. It is pointed out in a continuation that is to 

 appear in the Philosophical Magazine for September, 

 that a density of Eether of the order of only 10"', which 

 would not appreciably affect the inertia of matter, 

 would involve rigidity comparable with that of steel, 

 and thus permit transmission of magnetic forces by 

 stress ; this solid sether is, however, as usual, taken 

 to be freely permeable to the molecules of matter. 



kind of permanent nucleus that retains 

 around itself an sethereal field of physical 

 influence, such as, for example, a field of 

 strain. We can recognize the atom only 

 through its interactions with other atoms 

 that are so far away from it as to be prac- 

 tically independent systems ; thus our di- 

 rect knowledge of the atom will be con- 

 fined to this field of force which belongs to 

 it. Just as the exploration of the distant 

 field of magnetic influence of a steel mag- 

 net, itself concealed from view, cannot tell 

 us anything about the magnet except the 

 amount and direction of its moment, so a 

 practically complete knowledge of the field 

 of physical influence of an atom might be 

 expressible in terms of the numerical values 

 of a limited number of physical moments 

 associated with it, without any revelation 

 as to its essential structure or constitution 

 being involved. This will at any rate be 

 the case for ultimate atoms if, as is most 

 likely, the distances at which they are kept 

 apart are large compared with the diam- 

 eters of the atomic nuclei ; it in fact forms 

 our only chance for penetrating to definite 

 dynamical views of molecular structure. 

 So long as we cannot isolate a single mol- 

 ecule, but must deal observationally with 

 an innumerable distribution of them, even 

 this kind of knowledge will be largely con- 

 fined to average values. But the last half- 

 century has witnessed the successful appli- 

 cation of a new instrument of research, 

 which has removed in various directions 

 the limitations that had previously been 

 placed on the knowledge to which it was 

 possible for human effort to look forward. 

 The spectroscope has created a new as- 

 tronomy by revealing the constitutions and 

 the unseen internal motions of the stars. 

 Its power lies in the fact that it does take 

 hold of the internal relations of the indi- 

 vidual molecule of matter, and provides a 

 very definite and detailed, though far from 

 complete, analysis of the vibratory motions 



