450 SHOOTING STAES. [SECT, xxxvn. 



for those splendid exhibitions of falling stars which take 

 place at that season. 



The orbit already described is that which he formerly 

 assigned to this nebulous or cometary body, but he is now 

 of opinion that it has a period of something less than a 

 year, which would not only account for the shooting stars 

 of the 13th of November, but would also account for those 

 that happen at all seasons, and for some very great showers 

 of them that have taken place on two occasions near the 

 end of April. In the position assigned to this orbit by Pro- 

 fessor Olmsted, showers of shooting stars may happen in No- 

 vember and April. A very able memoir has been published 

 by M. Biot, in which that great philosopher shows that, in his 

 opinion also, meteoric showers are owing to the zodiacal 

 light coming into periodic contact with the atmosphere of 

 the earth. Which of these conjectures may be nearest the 

 truth, time alone can show ; but certain it is that the recur- 

 rence of this phenomenon at the same season for seven suc- 

 cessive years proves that it can arise from no accidental 

 cause. 



It is now well ascertained that great showers of shooting 

 stars occur also on the 12th of August, so that the earth's 

 atmosphere comes into contact with a zone of these small 

 bodies twice in the year. But, as shooting stars are seen 

 almost every night when the sky is clear, Sir John Lubbock 

 has thought it probable that some of these bodies may have 

 come so near, that the attraction of the earth has overcome 

 that of the sun, and caused them to revolve as satellites 

 round it. Should that be the case, they might shine by the 

 reflected light of the sun, and suddenly cease to be visible 

 on entering the earth's shadow. The splitting of the fall- 

 ing stars like a rocket, and the trains of light, may be ac- 

 counted for by supposing the stars to graze the surface of 

 the shadow before being eclipsed; and the disappearance 

 would be more or less rapid, according to the breadth of the 

 penumbra traversed. The calculations of M. Petit, Director 



