1 64 



LIGHT SCIENCE FOR LEISURE HOURS. 



not true either ol the August meteor system or of the 

 November one. Neither of these systems could be- 

 discernible at all during any total eclipse of the sun ; 

 for the simple reason that neither is rich enough 

 in meteoric matter nor illuminated with sufficient 

 brilliancy by the sun. We can be certain that such 

 meteor systems could not be seen during a total eclipse, 

 if we take into account a simple consideration which 

 Mr. Abbe seems to have entirely overlooked. If a 

 meteor system like the November system, which has 

 its greater part outside the orbit of the earth, could be 

 seen at all it would be seen during the darkness of 

 midnight. Let me be understood. I do not mean 

 that on every night, or even at any given season in 

 each year, we might expect to see the November or 

 August meteors. But this is certain, from the known 

 position and movements of these systems, that on 

 many occasions during the last century (to go no 

 further back) they should have been seen far more 

 conspicuously on the midnight sky that is, on the 

 side of the stellar sphere opposite to the sun than 

 they ever could be seen during total eclipse. In one 

 case we see them when we are nearer to them by a 

 whole diameter of the earth's orbit (or 185,000,000 miles) 

 illuminated fully like planets in opposition, and on a 

 dark background upon which stars down to the sixth 

 or seventh magnitude are visible to the naked eye. In 

 the other case, we look at them athwart the place of 

 the sun, and on a sky which, though it seems dark by 

 comparison with the brightness of mid-day, is yet very 



