On Peals of Thunder. 105 



undergo the same diminution. Let us cut ofF from the two 

 lengths placed in comparison in the preceding inequality, the 

 shortest visual ray, extending from the flash to the observer''s 

 eye, and there will remain, on the one hand, the difference be- 

 tween the longest and shortest visual ray, and on the other, the 

 short visual ray -f the length of the flash — the short visual ray, 

 or in other words precisely the length of the flash. It thus re- 

 mains established that the difference of the two visual rays in 

 question is smaller than the length of the flash of lightning. 

 When this difference can be valued in yards, we have then the 

 minimum limit for the length required. Let us now, then, in- 

 quire if the valuation of the difference of the two visual rays in 

 yards be possible. 



Why is the flash of lightning followed by a peal of thunder ? 

 Because its different parts are, in general, at unequal distances 

 from the observer. How long is the duration of the peal ? 

 This duration, we have already explained, is the time which the 

 sound requires to travel, an interval equal to the difference 

 of the length of the two lines extending from the two extremities 

 of the flash. Multiplying the number of seconds which the 

 peal has continued by 368.5, we have, in yards, the difference 

 of the two visual rays extending from the two extremities of the 

 flash, just as if it were possible to measure this difference in 

 space. The result of the multiplication will be the minimum 

 limit which we require. Example. — It has already been noticed 

 that M. de VIsle observed at Paris, in the year 1712, thunder, 

 whose peals lasted 39, 41, and 45 seconds. B}' multiplying 

 these three numbers by 368.5, we shall have respectively 

 14,371.5; 15,108.5; and 16,582.5 yards, that is to say, that the 

 corresponding flashes had a minimum length, 3.3 leagues ; 3.4 

 leagues ; and 3.8 leagues. Who would have expected these 

 enormous results ? 



For the purpose of definitely fixing our icfeas, I have sup- 

 posed, in the foregoing instance, that the flash was situated 

 wholly on one side of the zenith. Any other hypothesis, how- 

 ever, will in no degree affect the results we have obtained. The 

 calculated limits (for, wanting an angle, we have been able only 

 to discover the limits) will only be in these coses still farther 

 below the real length of the flash. 



