With respect to the observation of Dr Kirwan, that the barometer commonly falls after an aurora, this is no more than what takes place also after a thunder storm ; and its being followed by high winds from the south is as explicable on the electrical, as on the inflammable gas theory. Mr Winn, in the 73d volume of the Phil. Trans. makes the same remark, and says, that in 23 instances without fail, a strag gale from the south, or south-west, followed the stp pearance of an aurora. If the aurora were bright, he says the gale came on within 21. hours, but was of no long continuance ; if the light was faint and dull, the gale was less violent, and longer in coming on ; but longer also in duration. His observations were made in the English channel, where such winds are very dangerous, and by attending to the aurorae, he says, that he often escaped shipwreck, while others Buffered. As we have supposed a stream of electri city to be constantly Passing through the mass of the earth, from the equator towards the poles, it is evi dent, that a wind may be occasioned by this electri city finding a ready vent at some promontory, or head-land. And should we suppose one of those vents situated on the coast of France, or in the Bay of Biscay, the electric matter that has been received at the equator, during an aurora borealis, will be dis charged there, for some time after ; and consequent ly will occasion a wind from that quarter, which will be south-west in the English channel. According, however, to the different situations of these electrical vents, winds may blow in very different directions in different quarters of the world.
The most unaccountable of all the circumstances respecting the aurora borealis, is, that it is not much more than a century since this phenomenon has been observed to occur with any degree of frequency in our latitudes. We find indeed a few remarkable at mospheric phenomena recorded by the ancients, which may be reckoned examples of this meteor, viz. in Aristotle's Meteor. 1. i. c. 4, 5. ; and Senal. Quest. .Nat.. 1. i. c. 15. Pliny also (1. ii. c. 27.) speaks of a bloody appearance of the heavens, which seemed like a fire descending upon the earth, seen in the third year of the 107th Olympiad ; and of a light seen in the nightltime;equal to the brightness of day, in the consulship of Clecilius and Papyrus (1. ii. c. 33.) both which may be referred to the aurora borealis. But, with such trifling exceptions as these, the whole of antiquity is absolutely silent on this subject. Dr Halley informs us, that he had begun to despair of witnessing this beautiful phenomenon, when the re markable aurora of 1716 made its appearance. This philosopher has given us a historical detail of the se veral observations of this meteor, in which he says, that the first account of it upon record, in an English work, is in a book entitled, A Description of Meteors, by W. • F. D. D., reprinted at London, in 1654,
which speaks of burning spears" being seen Jan. 30. 1560. The next appearance of a like kind is re corded by Stow, and occurred on October 7. 1564. In 1574, according to Stow and Camden, an aurora was seen for two successive nights, viz, the 14th and 15th of November. The same phenomenon was twice seen in Brabant in 1575, viz. on the 13th of February and the 28th of September ; and the circumstances accompanying it wcte described by Cornelius Gem ma, who compares them to spears, fortified cities, and armies fighting in the air. In 1580 and 1581, this phenomenon was repeatedly observed at Baknang, in the county of Wirtemberg, in Germany. But from this time to 1621, we have no such phenome non on record, when it was seen all over France on September 2., and is particularly described by Gas iendi in his Phgsics, under the title of Aurora Bo realis. In November 1623, another was seen all over Germany, and is particularly described by Kepler. Since that time, for more than eighty years, we have no account of any such phenomenon being observed ; but in 1707, Mr Neve observed one of short conti nuance in Ireland ; and in the same year, a similar appearance was seen by Romer at Copenhagen ; while, during an interval of eighteen months, in the years 1707 and 1708, this sort of light had been seen no less than five times. The aurora of 1716, which Dr Halley particularly describes, was remarlmbly bril liant. It was also visible over a prodigiou7tract of country ; being seen from the west of Ireland to the confines of Russia, and the east of Poland ; extend, ing near 30° of longitude, and from abOut the 50th degree of north latitude, over almost all the north of Europe ; and in all places exhibiting, at the same time, appearances similar to those observed in Lon don.
It appears, then, to be certainly established, that the aurora was of very rare occurrence in our lati tudes till about a century ago ; for it cannot be sup , posed that so beautiful and striking a phenomenon would have passed unnoticed, and unrecorded, during ' the two preceding centuries, while men of science, and particularly astronomers, were so busily employed in examining every remarkable appearance of the heavens ; or that the philosophers of Greece and Rome would have remained silent concerning sd beau tiful a meteor, had it been in any degree familiarly known to them. It is in vain to account for their • silence by saying; that they inhabited latitudes which are scarcely ever visited by this appearance ; for the Romans not only visited, but long resided in the north of Germany, and in Britain, where the aurora, is now frequently seen in great splendour.