﻿in Total Eclipses of the Sun, 233 



we have already seen, was actually noticed by Professor Eastman. 

 It, of course, may happen that inequalities in the amount of 

 outstreaming on opposite sides of the equator may throw the 

 more prominent and conspicuous parts of the corona to the one 

 side or the other of the plane of the equator. It will be ob- 

 served that, from our present point of view, the extension of the 

 corona in the plane of the sun's equator is a phenomenon kindred 

 to the much greater luminous extension seen in the zodiacal 

 light — the only difference between them being that in the former 

 the auroral emanations proceed from lower heliographic latitudes, 

 and intersect nearer the sun. 



But it may be asked, how are we to explain, on the present 

 theory, the " star-points " of the corona over the polar regions 

 of the sun ? For these, two reasons may be assigned : — (1) If 

 we admit a distribution of magnetism on the sun similar to that 

 which prevails on the earth, the auroral streamers should diverge 

 from each other less rapidly in the high than in the low latitudes. 

 (2) Upon opposite sides of a line of no decimation traversing 

 the sun's surface, analogous to that which traverses Russia, the 

 natural directions of the streamers prolonged upward would be 

 such as to occasion the convergence and intersections of those 

 proceeding from the opposite sides of this line. 



We may say, then, that the more extended portions of the 

 corona, in the eclipse of 1869, were over those regions of the 

 sun's surface, and those only, where upon the present theory the 

 intersections of streamers might be expected to occur. 



In some eclipses distinct luminous curves having the appear- 

 ance of luminous jets issuing tangentially to the sun's limb, or 

 obliquely inclined to it, and pursuing a course either convex 

 or concave to the limb, have been seen. According to M. Liais 

 these peculiarities were conspicuously observable in the eclipse 

 of Sept. 7, 1858. While it is possible that such curves may be 

 the result of the intersections of a mass of straight streamers, 

 it is not improbable that they may be actual luminous jets ; for 

 if from any cause any portion of the auroral matter should be 

 projected from the sun in a direction oblique to the surface, it 

 would proceed in a convex hyperbolic curve if repelled by the 

 sun, and in a concave curve if attracted. Now I have shown, in 

 a former number of Silliman's Journal (July 1861), that the por- 

 tion of cometary matter posited on the convex side of the tail 

 of Donati's Comet was actually repelled by the sun, while that 

 on the concave side had become detached from the head of the 

 comet, because of a diminished gravitation toward the sun. Upon 

 our fundamental conception that the coronal matter is essentially 

 in the same physical condition as such cometic matter, and sub- 

 j ect to the action of the same solar forces, it may well happen 



