114 CELESTIAL PHENOMENA. 



confirmed by Peldt's careful calculations, shewing that, for 

 want of perfect simultaneity in the observed disappearances, 

 an upward movement cannot be regarded as a result of 

 observation ( 70 ). Future researches must decide whether, 

 as Olbers supposes, the explosion of shooting stars, and the 

 ignition of fire-balls, may not occasionally impel the 

 meteors upwards, or otherwise influence the direction of 

 their paths. 



Shooting stars fall either singly or sporadically, or in 

 groups of many thousands which are compared by Arabian 

 writers to flights of locusts. The latter cases are periodical, 

 and the meteors are then seen in streams, moving, for the 

 most part, in parallel directions. Of the periodic groups, 

 those hitherto best known are the phsenomena of the 12th 

 to the 14th of November, and of the 10th of August or the 

 day of St. Lawrence, whose "fiery tears'' ( 71 ) were long since 

 recognised in England as a recurring meteorological phse- 

 nomenon, and are mentioned in an old Church Calendar, as 

 well as in legendary traditions. Although a mixed shower 

 of shooting stars and fire-balls had been seen on the night 

 of the 12th and 13th of November, in 1823, by Kloden, at 

 Potsdam, and in 1832 throughout Europe, from Ports- 

 mouth in England, to Orenburg on the Oural river, and even 

 in the southern hemisphere in the Isle of Prance, still the 

 idea of the periodicity of the phsenomenon, and of great 

 showers of falling stars being connected with particular days, 

 was first inferred on the occasion of the fall observed by 

 Olmsted and Palmer in North America on the 12th and 

 13th of November, 1833, when the shooting stars seemed, 

 in one part of the sky, to fall as thickly as snow-flakes ; 

 in the course of nine hours there fell at least 240000. 



