66 WRITINGS OF JOSEPH HENRY. [1832 



mus Hall, on Long Island. It also appeared brilliant at 

 Potsdam in St. Lawrence county, the most northern Acad- 

 emy in the State. It was probably not seen very extensively 

 in the States east of New York, as I am informed the weather 

 in the eastern part of New England was cloudy at the time, 

 accompanied with rain. The aurora is described as shoot- 

 ing up to the zenith at North Salem ; and at Middlebury as 

 consisting of coruscations in almost every part of the visi- 

 ble heavens. At Fairfield, it illuminated nearly the whole 

 heavens; a number of bows, commencing in the northwest, 

 passed south of the zenith, and terminated in the northeast. 

 An interesting account is given of its appearance at Utica, 

 where it is described as rising at one time in streams of 

 light, of purple, yellow, green, and other colors, and exhib- 

 iting a rapid horizontal motion, passing and repassing like 

 a company of dancers. The actual intersection of the beams 

 so as to form the appearance called the corona, is mentioned 

 as having been seen in the city of New York, at Hartwick, 

 Cherry Valley, Hudson, and Prattsburg in Steuben county. 

 The only plausible explanation of the formation of the 

 corona, is that which supposes the beams of the aurora to 

 consist of cylindrical portions of some kind of matter, which 

 becomes luminous as it passes into the higher regions of the 

 atmosphere; and that the cylindrical beams shoot up from 

 many points of the earth's surface, nearly parallel to each 

 other, and in the direction of the dipping needle. Being at 

 different distances from the observer, they appear of differ- 

 ent elevations; and sometimes, when seeming to overlap 

 each other, they form continued streaks of light in every 

 part of the visible heavens. The corona, according to this 

 hypothesis, is the perspective projection on the sky, of the 

 beams which are shooting up at the same instant on all 

 sides of the observer, and which, being all parallel to the 

 dipping needle, appear to converge as it were to a vanishing 

 point, situated, in the State of New York, about 15° south 

 of the zenith. If this hypothesis be correct, (and it seems 

 a strict geometrical deduction from actual appearances,) it 

 would follow that on the evening of the 19th of April, 



