PART II. POLAR MAGNETIC PHENOMENA AND TERRELLA EXPERIMENTS. CHAP. V. 631 



graphs of these registerings are given, which show that corresponding small sudden alterations in 1) 

 were simultaneous within three seconds in Potsdam and Bossekop. 



According to my theories of magnetic storms, it might be expected that sudden similar magnetic 

 ,-lianges which occur in different parts of the earth arise rather simultaneously. When the sun suddenly 

 nds forth a strong pencil of cathode rays towards the earth, this pencil, owing to earth-magnetism, will 

 ae broken up in such a way as to form different partial systems of magnetic impulses -- polar and 

 equatorial. The various groups of rays have to travel different way-lengths in space before reaching 

 :hrir nearest to the earth, and may arrive at very different regions of the earth for the different groups. 

 But the difference in time between the various impulses affecting any particular locality on the earth 

 ran scarcely be more than a couple of seconds, while the difference in the intensity of the effects can 

 je very considerable. We know of corresponding phenomena in the case of Aurora, which will be 

 .reated later on. 



COMETS' TAILS. 



124. The theory here set forth, of the emanation of electrical corpuscle-rays from the sun, might 

 be thought to present a new point of departure in the study of the physical nature of comets, and more 

 (.specially of comets' tails. 



It seems evident from their spectrum that comets consist of an accumulation of cosmic dust, with 

 various carbonaceous substances, concentrated about one or more nuclei, which are surrounded by a 

 highly rarefied vaporous envelope in which possibly carbonaceous gases are comparatively strongly 

 represented. 



As regards more especially the particular phenomenon of the comet's tail, it has been found that 

 it docs not make its appearance until the comet aproaches the sun, and is most highly developed a little 

 vvhile after passing the perihelion. 



If, now, this vaporous envelope surrounding the more solid part of the nucleus, be ex- 

 losed to the radiation of a multitude of corpuscle-rays from the sun, it could easily be imagined 

 .hat in their passage through the exceedingly rarefied gas, these rays would change their nature. The 

 simplest assumption one is inclined to make is that some of the corpuscles that pass through the coma 

 lave acquired an appendix of gaseous atoms or molecules, which have thereby become luminous. As 

 .hese rays may be supposed to continue their way in more or less the same direction as before, but 

 A-ith a different velocity and mass, this would be a comparatively simple explanation of the luminous tail 

 if the comet, which is almost always directed away from the sun. 



AKRHKNIUS has also, as we know, maintained a similar theory, only that instead of electric corpuscle- 

 ays of the kind here considered, he imagines rays of electrically-charged atoms, moving under the 

 nfluence of light-pressure. 



It is possible, however, that there are also other, just as natural, ways of looking at the matter. 

 It might be imagined that after great heating by direct insolation, the comet is charged negatively by 

 -athode-rays from the sun, and that the charging reaches so high a potential that the comet dis- 

 :harges itself electrically, so to speak in the direction of its own shadow. These discharges may also 

 )e imagined to be due to some extent to an emission of secondary rays from the cosmic dust of the comet. 



I have been led to this thought by experimental analogies which will be described farther on. 

 Answering to the idea that a comet is an accumulation of carbonaceous cosmic dust almost without atmo- 

 sphere, I have carried out experiments in which the cathode in a vacuum-tube consisted of a carbonaceous 

 naterial. The most recent investigations of the comet-spectrum seem to indicate that the radiation from 

 i comet may be compared to that from a cathode in a Crookes' tube (DESLANDRES, FOWLER). 



