304 



NATURE 



[July 30, li 



in properties from ordinary light that it required the genius o 

 Clerk-Maxwell to recognise them as light at all. I shall have 

 to call your attention to another kind of light, discovered lately 

 by Becquerel. It will doubtless be urged that widely as these 

 kinds of light difi'er from each other in some respects, they all 

 are bent when they pass from one substance to another, while 

 the Riintgen rays, as we have seen, are not refracted. This 

 objection to the possibility of the Rontgen rays being a kind of 

 light, formidable as it appears at first sight, loses all its force 

 when more closely examined. We know cases in which light 

 passes through substances without being refracted ; thus Kundt 

 found that certain rays could pass through gold without being 

 refracted, while other rays were bent the wrong way. Stenger 

 has lately found that certain blue rays can pass through fuchsin, 

 and other slightly different ones through Hofmann's violet, 

 without being bent. Perhaps, however, the most striking 

 testimony to there being nothing inconsistent in the idea of a 

 kind of light which is not refracted, is afforded by one of the 

 last investigations undertaken by von Helmholtz, and published 

 about three years ago. Von Helmholtz investigated what, on 

 the electromagnetic theory of light, would be the bending 

 experienced by light of different frequencies passing through an 

 ideally simple substance, one whose spectrum consists of only 

 one line. The result of his investigation is shown in this curve 

 (Fig. 2),where the abscissa; represent the frequency of the light, the 

 ordinates the refractive index. On the part of the curve from 

 / to g, you see that the refractive index increases as the frequency 

 increases ; this corresponds to the normal spectrum where the 

 blue rays are more refracted than the red. After passing b the 

 curve dips down : this means that the greater the frequency the 

 less the bending, in other words, the blue rays tend to be 

 bent more than the red. We know many instances of 

 this, it is called anomalous dispersion. Then we get to 



the part of the curve about i, where the refractive index 

 is less than i ; that is, where the rays are bent the wrong 

 way. We have examples of this, as Kundt has shown, in the 

 case of gold, silver and copper ; but the most interesting part of 

 the curve for our purpose is the last part, where, after dipping 

 below the line of no-bending for a short distance, it approaches 

 it and practically coincides with it for all frequencies greater 

 than a h ; so that, on this theory there would practically be no 

 bending for all waves whose frequency exceeds a certain value. 

 Thus, so far from the absence of bending being a proof that the 

 Rontgen rays are not light, this absence of bending is exactly 

 what we should expect if these rays were light of very great 

 frequency. 



A characteristic feature of all varieties of light is the exist- 

 ence of polarisation, and polarisation is indisputable evidence 

 of transversal vibration ; hence, many experiments have been 

 made to see whether any polarisation of these rays could be 

 detected. All these experiments have practically been confined to 

 seeing whether the Rontgen rays could traverse two plates of 

 tourmaline more freely when the axes of the two plates are parallel 

 than when they are crossed ; there is a great difference in the 

 transparency to ordinary light in the two cases. The results of 

 these experiments are somewhat conflicting. Prince Galitzine 

 and M. de Karnojitsky are of opinion that they have suc- 

 ceeded in detecting a slightly greater absorption of the rays 

 when the axes are crossed than when they are parallel ; on 

 the other hand, Becquerel, Mayer, and I were not able to detect 

 any appreciable difference in the two cases. If the result of 

 Prince Galitzine should be confirmed, it would prove beyond 

 cavil that these Rongten rays were light ; but even if the 

 presence of polarisation is not definitely established in this case, 



NO. 1396, VOL. 54] 



it does not follow that these rays can not be polarised — the 

 methods for polarising one kind of light may not be successful 

 when used for another. For example, a wire bird-cage will 

 polarise the long electrical waves, but will not affect the shorter 

 waves of radiant heat, much less those of visible light. By wind- 

 ing exceedingly thin wires close together on a framework, 

 Rubens and Du Bois were able to polarise the waves of radiant 

 heat, the wave-lengths of which are long compared with those ol 

 light. This arrangement, however, is much too coarse to polarise 

 visible light, much less ultra-violet light. And it is possible, and 

 indeed likely, that the structure of the tourmahne, though fine 

 enough to polarise ordinary light, may not be fine enough to 

 polarise the Rontgen rays. 



So far, I have confined myself to showing that there is 

 nothing in the effects known to be due to these rays inconsistent 

 with their being a variety of light. I must now pass on to 

 some evidence of a more positive character. Since the discovery 

 of the Rontgen rays, Becquerel has discovered a new kind of 

 light, which in its properties resembles the Rontgen rays more 

 closely than any kind of light hitherto known. Becquerel 

 found that certain uranium salts emitted, after being exposed to 

 the sunlight, radiations which, like the Rontgen rays, could pass 

 through plates of aluminium or of cardboard, and affect a photo- 

 graphic plate behind. I have here a photograph of a perforated 

 piece of zinc, which has been taken by Becquerel's method by 

 simply scattering over the zinc plate powdered uranium nitrate, 

 and placing it over a photographic plate well protected from 

 ordinary light. After a long exposure of from twenty to forty 

 hours, the photograph now on the screen was taken. Becquerel 

 has shown that the radiation from the uranium salts can be 

 polarised, so that it is undoubtedly light ; it can also be refracted. 

 It forms a link between the Rontgen rays and ordinary light, it 

 resembles the Rontgen rays in its photographic action in power 

 of penetrating substances opaque to ordinary light, and in the 

 characteristic electrical effect, while it resembles ordinary light in 

 its capacity for polarisation, in its liability to refraction. The 

 persistence of the radiation is very remarkable. Becquerel 

 found that the potassium-platinum compound of uranium went 

 on emitting these radiations with nearly undiminished zeal for 

 fifteen days after it had been exposed to the sunlight. It would 

 seem that under the influence of sunlight some change in the 

 chemical or' physical nature of the substance occurred, and that 

 after the sunlight was cut off, the substance gradually went 

 back to its original state, and that while doing so it emitted this 

 peculiar radiation. The radiation from the uranium salts is of 

 especial interest from another point of view. Sir George 

 Stokes has shown that in the case of phosphorescence caused 

 by sunlight or the arc lamp, the light emitted by the phosphor- 

 escent body is of longer wave-length than the light causing the 

 phosphorescence ; in the case, however, of the phosphorescence 

 discovered by Becquerel, the light emitted is of a shorter wave- 

 length than the incident light. The case resembles that called 

 calorescence by Tyndall, when the body placed in a focus of 

 dark radiant heat becomes luminous and gives out the shorter 

 luminous waves. 



From this discovery of Becquerel, we may conclude that 

 besides the vibrations emitted by luminous bodies with which 

 we have hitherto been acquainted, there are others having a 

 much greater frequency and, it may be, arising in a different 

 way. 



To sum up, we may say that though there is no direct evi- 

 dence that the Rontgen rays are a kind of light, there is no 

 known property of these rays which is not possessed by one or 

 other of the forms of light. 



One of the most remarkable phenomena connected with 

 these rays is the way in which the absorption depends upon 

 the density of the body ; if we measure the transparency 

 of a series of bodies, we find that the order of opacity is the 

 same as the order of their density. No other factor in the consti- 

 tution of the body seems comparable in importance with density. 

 In this respect, the relation between the opacity and the 

 other properties of a body in the case of the Rontgen rays is 

 simpler than that for luminous waves or electric waves. There 

 seems no simple relation between the density of a body and 

 its transparency to visible radiation or electrical vibration ; in the 

 case of the Rontgen rays, however, it seems the greater the 

 density the greater the opacity. This appears to favour Prout's 

 idea that the different elements are compounds of some prim- 

 ordial element, and that the density of a substance is proportional 

 to the number of the primordial atoms ; for if each of these 



i 



