No^'EJrBER 1, 1894.] 



KNOWLEDGE. 



i)0 



predominant, hence definite lines in the spectrum of each 

 kind of molecule. I further picture these molecules flying 

 about and colliding. The greater the vibration of the 

 molecules, the higher the velocity in their respective paths ; 

 and the greater the velocity, the greater the collisions and 

 consequent vibration. Thus give your molecules an 

 impulse of vibration, or of translation, and the result in a 

 short time is the same. 



I can confirm Mr. Evershed's experiment with iodine 

 vapour, in part at least, although I performed the experiment 

 from another point of xievf. I imagined that if an absorb- 

 ing vapour were viewed from the side, a spectrum, 

 complementary to that seen by transmission, should appear. 

 To this end, I enclosed iodine in a small cage of wire gauze 

 with apertures at right angles to each other. I passed the 

 rays from a lime-light through in one direction, focussing 

 the image of the lime in the cage so as to get the most 

 intense illumination, and viewed the transmitted light. 

 The iodine was then vaporized by a flame below and the 

 usual absorption spectrum viewed, and simultaneously 

 the vapour was examined by a spectroscope pointed at 

 right angles to the former. I saw no complementary 

 spectrum, but only a very faint general illumination, too 

 faint to call a continuous spectrum. The temperature was, 

 however, low. I put my failure to see a complementary 

 spectrum down to the probable fact that the vibrations set 

 up by absorption, already faint, would be communicated 

 to molecules outside those immediately absorbing from the 

 lime-light on their way to the eye, and so the vibrations 

 would be frittered down to too small an extent to be 

 visible. 



The faint general illumination seen by me, and the 

 continuous spectrum seen by Mr. Evershed, were probably 

 due to the closeness of the molecules, which in such case 

 have not suificient time between collisions to settle down 

 into their natural periodic vibrations, as in the case of a 

 solid or liquid. Many years ago. Prof. Lockyer showed 

 the widening of the lines of hydrogen by increase of 

 pressure, and Mr. Evershed's experiment with sodium 

 vapour points to this explanation. 



Owing to the large number of diflerent periods in which 

 the iodine molecule can vibrate, as evidenced by its 

 absorption spectrum, it would appear probable that a less 

 density of the vapour under examination would be necessary 

 in order to give its specific spectrum than in the case of 

 sodium, with its more simple vibrations. 



Yours truly, 



Rugby. G. M. Seabroke. 



[Mr. Seabroke's experiments on the radiation from 

 vacuum tubes electrically excited appear to show, as he 

 says, that there is no essential difl'erence in the radiation 

 of a gas, whether it is produced electrically or otherwise. 

 It is probably entirely a question of temperature, or, for 

 any given substance, molecular velocity, and it appears to 

 me that all ordinary cases of gaseous radiation can be 

 explained in this way without resorting to the hypothetical 

 phenomenon called " luminescence." Thus Prof. Smilhells 

 has shown that, in the case of flames tinted with metallic 

 salts, where the radiation exceeds that of a perfectly black 

 body raised to the average temperatiu-e of the flame (for 

 the rays special to the metal or salt), this so-called 

 " luminescence " of the salt may be really due to the fact 

 that the arenvje temperature is very much lower than the 

 temperature of such molecules in the flame as are in the act 

 of combining ; and that the intense radiation, therefore, may 

 after all be merely a heat effect, the extremely high tempera- 

 ture of the comlsining molecules being communicated to a 

 certain proportion of the salt molecules. With regard to 



the iodine radiation, I do not think it is simply a question 

 of molecular aggregation acting in the way Mr. Seabroke 

 supposes. I have observed the iodine emission spectrum 

 under a variety of conditions as to density, but in no case 

 can I get any indication of a discontinuous emission, and 

 think that this can only be obtained with higher tempera- 

 tures than I have hitherto employed. — J. Evershed.] 



GLOBULAR STAR CLUSTERS. 

 To the Editor of Knowledge. 



Dear Sir, — With reference to your remarks on my paper 

 in the October number of KNn\vLED(iE, in which you say 

 that my views carry us "back to the old theory, that 

 clusters and nebulfe are distant galaxies, outside and un- 

 connected with the Milky Way," I would like to say that 

 I do not believe that the clusters and nebulae are external 

 galaxies. On the contrary, I say in my paper, " Judging, 

 however, from the average distance recently found for stars 

 of the first and second magnitudes, the distance of ordinary 

 stars of the fourteenth magnitude — on the Supposition 

 that they are of the same size and brightness, and that 

 their light is simply reduced by distance — would be about 

 ten times greater than that found above for Omega 

 Centauri." This, I think, clearly implies that I consider 

 the globular clusters to be included within the limits of 

 our sidereal system. If any external universes exist, 

 which seems probable, I do not think that any of them are 

 within the range of our largest telescopes. These views I 

 have already expressed in my " Visible Universe." 



That some of the brighter stars may be many times 

 larger than the sun I admit, but that bodies of the size of 

 the earth should retain for ages their solar brilliancy 

 seems to me very improbable. Some of the fainter stars 

 in the I leiades may lie far beyond the cluster itself. 



Yours faithfully, 



J. E. Gore. 



[The way m which star clusters are distributed in the 

 heavens along the stream of the Milky Way, shows that 

 star clusters are intimately associated with the Milky Way, 

 and that they are at about the same distance from us as 

 the nebulous stream. Herschel's " flat grindstone theory" 

 of the structure of the Milky Way has, I think, now been 

 very generally abandoned by modern astronomers, and the 

 generally received idea is that the Milky Way stream has 

 probably a roughly circular section, so that the most 

 distant stars in the stream are probably not fifty per cent, 

 more distant from us than the nearer stars in the stream. 



If one iDart of the nebulous stream which encircles the 

 heavens were ten times as far removed from us as another, 

 we should expect to find the more distant part appearing 

 narrower or fainter by reason of its distance ; but there is 

 no very great ditl'erence in the general breadth or bright- 

 ness of diflerent parts of the Milky ^^'ay stream. It seems 

 therefore probable that the nebulous ring is not very far 

 from circular, and that the earth is not very far removed 

 from the centre of the ring ; that is, one part of the stream 

 is probably not two or three times as far removed from us 

 as another. We cannot, therefore, assume that star clusters 

 which are closely associated with the nebulous stream, or 

 are distributed along its borders, are some of them ten 

 times as distant from us as others. 



Mr. Gore's reasoning with regard to the smallest con- 

 ceivable size of luminous star depends upon the assumption 

 " that they retain for ages their solar brilliancy" ; but man 

 has only observed them for a few years, and we have no 

 evidence that they are not rapidly cooling and were not 

 raised to incandescence by a comparatively recent collision. 

 — A. C. Kanyard.] 



