THUNDER-STORMS. 



525 



Mariners state that thunder heard at sea is marked by rolling as loni; con- 

 tinued as on land, although none of those causes which are generally supposed 

 to produce echoes, such as walls, rocks, wood, hills, or mountains, are present. 

 Unless the surface of the clouds reflects sounds, no means of producing an echo 

 can exist under such circumstances. Although it might seem that the clouds 

 would be as little capable of reflecting sound as the air itself, there appears to 

 be some reason to judge otherwise. Muschenbroeck states, as the result of 

 his own observations, that a cannon, which, being discharged when the heavens 

 are unclouded, produced only a single report, had its sounds several times re- 

 verberated when discharged in the same place under a clouded sky. in the 

 course of the experiments made in 1822, to determine the velocity of sound al- 

 ready referred to, the same observation was made. 



In the posthumous works of Hooke, published in 1706, an explanation was 

 proposed for the rolling of thunder, which was more recently reproduced with 

 more full developments by Dr. Robinson in the Encyclopaedia Britannica, and 

 ) which seems more adequate, and open to fewer objections, than any other hy- 

 ^ pothesis yet suggested. The sound is supposed to be developed by the light- 

 ; ning in passing through the air, and consequently separate sounds are pro- I 

 1 duced at every point through which the lightning passes. As the object of the / 

 hypothesis is to explain the rolling or succession of sounds, and not the sound ^ 

 itself, it is immaterial what the manner of producing the sound may be. 



Let us first suppose that the lightning were to move in a circle, of which the 

 observer is the centre. The velocity of the lightning is so extreme that, for 

 the purposes of this explanation, it may be assumed to be at the same moment 

 in every part of the circle. Explosions will, therefore, be produced simulta- 

 neously at every point in the circumference of the circle, and, as all these 

 ) sounds have the same distance to traverse in coming to the observer, they will 

 ) arrive at his ear at the same instant ; the effect would, therefore, be a single 

 \ sound, having a force due to the combined effects of all the sounds produced 

 ; in the circumference of a circle. To apply this reasoning to the actual case 

 of thunder, let it be supposed that two small clouds oppositely electrified are 

 situated near each other, and at the same height in the zenith of the observer. 

 The clouds may be considered as placed in the surface of a sphere, in the cen- 

 tre of which the observer stands. If the electric discharge takes place between 

 the clouds, the thunder would be heard by the observer as a single clap, with- 

 out any roll or reverberation. 



Let us next suppose the lightning to move in any line which is not part of a 

 circle or sphere, with the observer in the centre ; let its course be a straight 

 line, for example, such as A B, the observer being at O. From 0, suppose a 



Fig. 3. 



perpendicular, O L 1 , drawn to A B, and let two lines, L 2 , the lergth of 

 which shall exceed O L 1 by one hundred and ten feet, be inflected from O on 



