632 PHILOSOPHICAL TKAN8ACTI0NS, [aNNO l/QO. 



Philos. Trans, there are 2 resembling so exactly the above, that they deserve the 

 consideration of the learned ; one was seen in 1734-5, the other in 1749. 



Some years before also, Mr. P. observed a few others, very similar to that just 

 described ; he therefore adds a short account of them, viz. at Brussels, March 

 14, 1774, at about 7 o'clock in the evening, the sky being very clear, there ap- 

 peared an arch resembling a bright white fog, about 8 or 9° broad, tolerably 

 well defined ; the brightness of the stars it covered was diminished. The phe- 

 nomenon lasted about -| of an hour. The air was cold, but not frosty. To- 

 wards midnight an aurora borealis was seen in the north, which appeared some- 

 thing like the phenomenon just mentioned. Again, March 15, 1774, at about 

 74- o'clock in the evening, a column of light appeared in the north, something 

 like that of yesterday ; weather very fine. Also, at Louvain, 1775, April 19, 

 9^ 30"™, at night, after a storm, he saw a bright white line of light 1 or 2° in 

 breadth, extending from n. n. e. through n. to n. w. almost parallel to the hori- 

 zon, and elevated about 9°. It was brighter in the centre, and stars of the 3d 

 and 4th magnitude which it covered were much diminished in brightness ; it 

 sometimes rapidly vanished and re-appeared, and altogether lasted near half an 

 hour. Lastly, at Wickhill in Gloucestershire, 1777^ Feb. 26, at about 7*^ at 

 night, he saw a faint white tract of light, not unlike a foggy column, about 6 

 or 8° in breadth. It extended from the horizon w. by s. to e. by s. passing over 

 the stars in Orion's feet, and a very little to the north of Sirius. It seemed to 

 have no motion, or to alter in brightness. The air was rather foggy, with a few 

 clouds and a little wind. At about 10 o'clock a slight aurora borealis appeared 

 in the north with streaks, extending sometimes to the zenith. 



These kinds of lights seem to differ from the common aurora borealis in 

 several particulars : their light is more condensed ; they assume the form of an 

 arch or column, and appear either to the north or south of the zenith, though 

 he thinks oftenest to the south. 



VIII. Experiments on the Analysis of the Heavy Iriflammahle Air. By Wm. 

 Austin, M. D., Fellow of the College of Physicians, p. 51. 

 In a paper read before the r. s. in the year 1788, Dr. A. suggested an idea, 

 that the heavy inflammable air is a compound of the light inflammable and 

 phlogisticated airs. At that time he had observed, that the heavy inflammable 

 air, or at least fixed air, is formed on the decomposition of nitrous ammoniac by 

 heating it in close vessels ; and that this air is affected by the electrical shock, 

 like other elastic fluids into whose composition the light inflammable air enters. 

 The conclusion he then drew from those facts seems to be supported by several 

 subsequent experiments, which he here lays before the r. s. Several elastic fluids 

 containing the light inflammable air, as the hepatic and alkaline airs, being de- 



