SHOOTING STARS. 585 



Whether anything has ever fallen from the shooting-stars 

 to the Earth, has been much discussed in opposite senses. 

 The straw roofs of the parish Belmont (Departement de 1' Ain, 

 Arondissement Belley) which were set on fire by a meteor in 

 the night of November 13th, 1835, just at the epoch of the 

 known November phenomenon, received the fire, as it ap- 

 pears, not from a falling shooting- star, but from a bursting 

 tire-ball, which problematical aerolite is said to have fallen 

 according to the statements of Millet d'Aubenton. A similar 

 conflagration, caused by a fire-ball, occurred on the 22nd of 

 March, 1846, about three o'clock in the afternoon, in the 

 commune of St. Paul, near Bagnere de Luchon. Only the 

 fall of stones in Angers (on the 9th of July, 1822,) was 

 ascribed to a beautiful falling star seen near Poitiers. This 

 phenomenon, not sufficiently described, deserves great atten- 

 tion. The falling stars resembled entirely the so-called 

 Roman candles used in fireworks. It left behind it a straight 

 Ktreak, very narrow above, and very broad below, which lasted 

 for ten or twelve minutes with great brilliancy. Seventeen 

 miles northwards of Poitiers an aerolite fell with a great 

 detonation. 



Does all that the shooting-stars contain, burn in the outer- 

 most strata of the atmosphere whose refracting power causes 

 the phenomenon of twilight? The above-mentioned various 

 colours, during the process of combustion, admit of the in- 

 ference of a chemical difference in the substances. In addi 

 tion to this, the forms of these fiery meteors are exceedingly 

 variable ; some form merely phosphorescent lines of such fine 

 ness and number, that Forster, in the winter of 1 832, saw 

 the sky illuminated by them with a feeble glow.* 8 Many 

 shooting-stars move merely as luminous points, and leave no 

 tail behind them. The combustion, attended with rapid or 



* Fors tor's Mc'moire sur les Etoilesfil antes, p. 31. 



