AURORAS. 



177 



present themselves : that of the earth's atmos- 

 phere, first, is one in which the curvature of 

 the course of a ray passing through it is less 

 than the curvature of the globe such atmos- 

 phere surrounds. The moon having, if any. an 

 atmosphere rarer than that of the earth, the 

 like result must, a fortiori, hold of any atmos- 

 phere it may have. As a practical consequence, 

 the apparent diameter of the moon, as ascer- 

 tained by measurement, will be greater than 

 that interred from observation of an occulta- 

 tion of a star ; since, through bending of the 

 light by the moon's atmosphere, the star will 

 disappear and reappear when the line of vision 

 is a little within the moon's apparent boundary. 

 A like result would be obtained from a solar 

 eclipse. The author stated that by actual com- 

 parison of the two kinds of determination, such 

 an excess to the amount of 6" to 8" was found. 

 This difference, accordingly, he considers as 

 indicating a lunar atmosphere of very small 

 extent and density. The reasoning leads us to 

 anticipate that, in a solar eclipse, a slender 

 band of the sun's disk, immediately contiguous 

 to the moon's border, will appear somewhat 

 brighter than parts more distant : attention to 

 this feature at the next solar eclipse was ad- 

 vised. The other case, that in which the cur- 

 vature of path of the ray is greater than that 

 of the globe having the atmosphere, was as- 

 sumed true in the instance of the sun. On this 

 supposition, it was shown that all objects seen 

 by rays coming from the sun's periphery, 

 would by the great refraction be brought out 

 apparently to the level of the boundary of the 

 solar atmosphere; and this would be so, 

 whether the rays proceeded from objects on 

 the surface of the interior globe, or from cloud- 

 like masses suspended in its atmosphere. The 

 contour of the sun, accordingly, should appear 

 quite continuous, and the angle of its apparent 

 semidiameter will equal the angle subtended 

 at the earth by the whole height of the solar 

 atmosphere. For like reasons, the apparent 

 diameter of the planets will be augmented to 

 a certain amount by the effect of refraction ; 

 and on account of the great distances of these 

 bodies from the earth, the eclipse of a satellite 

 will take place as soon as the visual ray begins 

 to be bent by interposition of the atmosphere 

 of the planet. 



AURORAS. The questions relative to the 

 precise manner of production of the auroral 

 display, and its elevation and place with respect 

 to the earth, have be^en during the year 1862 

 somewhat pointedly discussed. The aurora 

 borealfe was, of course, known to the civilized 

 world earlier than its counterpart of the south- 

 ern polar regions; and the theories at first 

 proposed to account for it, among them that 

 which referred the light to the bending of the 

 sun's rays by the atmosphere about the north- 

 ern convexity of the earth, and that which sup- 

 posed it to be the effect of rays reflected from 

 icebergs, were either imperfect on physical, or 

 impossible on mathematical grounds, or defec- 



VOL. IL-12 



tive in both these particulars. Since the phe- 

 nomena of electricity became better known, the 

 view proposed by Cavendish, to the effect that 

 the aurora is due to the passage of electricity 

 through portions of the atmosphere in which 

 the air is considerably rarefied, or at all events 

 serves as a conductor, and yet not a perfect one, 

 has become (so far as it extends) quite generally 

 adopted. The causes and conditions leading to 

 disturbance of the electric equilibrium and to 

 the actual discharge, remained to be investiga- 

 ted. Many years since, it began to be observed 

 that not only are the instruments at the earth's 

 surface employed to indicate electric distur- 

 bance (electrometers) subject to excitation dur- 

 ing the auroral displays, but also that at such 

 times marked deviations or irregularities of the 

 magnetic needle almost invariably occur. At 

 tunes when the aurora has appeared simul- 

 taneously over a great extent, as in 1831, 1839, 

 and 1859, the magnetic disturbance has been 

 unusually great. It is now known also that 

 generally the display occurs at the same times 

 at both poles, or as aurora borealis and austra- 

 lis ; though it is believed that, for some reason, 

 the phenomenon is somewhat the more frequent 

 in and near the arctic zone. Of controversial 

 papers on the subject during the year, the most 

 important have been that of Prof. A. De la Rive, 

 read before la Societe de Physique et cTHistoire 

 yaturelle, of Geneva, Feb. 6, 1862, an abstract 

 of which appears in the " Philosophical Maga- 

 zine," vol. xxiii, p. 546; and that of Dr. David 

 Walker, in the " Intellectual Observer/' Xov. 

 1862. The chief addition to our physical know- 

 ledge of the subject is probably to be found in 

 the " Ninth Article " upon electrical currents 

 near the earth's surface, and their connection 

 with the aurora, by Prof. E. Loomis, of Yale Col- 

 lege, " Amer. Journal of Science," July, 1862. 

 Of course, no theory of the aurora can be 

 correct, unless according with the fact of the 

 actual position in space of the luminous arc or 

 streamers : this being a point not easily deter- 

 mined, and probably not yet known, each 

 theory must assume for the "display some posi- 

 tion to which its requirements will correspond. 

 Cavendish supposed the height of the aurora 

 about 71 miles, the air, if there existing at all, 

 having but T To.'r5Tr tn P art of its density at the 

 earth ; Halley, Dalton, and others assumed 

 heights of 52 miles or upward. The actual 

 height of the base of the visible arc should be 

 ascertained by trigonometrical measurement 

 from different positions at the same time ; yet 

 observers at different places must often see at 

 the same moment each a different aurora ; and 

 in an instance of measurement by two observ- 

 ers and supposedly of the same aurora. Jan., 

 1831, one of these made the height 18 miles, 

 the other not less than 96. Prof. Loomis, in 

 his " Eighth Article " npon auroras, especially 

 the great aurora of Aug. 28, to Sept. 4, 1859 

 (" Amer. Jour, of Science," Nov. 1861), by com- 

 paring the angles of elevation of the same arches 

 as seen from different positions, in a few in- 



