THUNDER-STORMS. 527 



flash, according to its direction with respect to the observer, being susceptible 

 of an infinite variety of sonorous effects. 



An objection to this fascinating hypothesis occurs to me, which appears to 

 have escaped the attention of its advocates, and which, nevertheless, is entitled 

 to consideration. I have supposed, for the sake of illustration, in the prece- 

 ding developments that a succession of distinct sounds are emitted at points of 

 a space the difference of whose distance from the observer is one hundred and 

 ten feet, and therefore these sounds succeed each other at intervals of a tenth 

 of a second. Any other difference of distance would equally serre the purpo- 

 ses of illustration, the interval between the successive detonations being deter- 

 mined by it according to the known velocity of sound. But it does not appear 

 to me that there is anything in the physical effects to warrant the supposition 

 of a series of separate sounds emitted at points of space more or less distant 

 from each other. The electric fluid rushes through space, producing the sarru 

 efft-cl at every point. The analogy on which Dr. Robinson bases the expla- S 

 nation (to a file of soldiers, placed at certain distances asunder, who discharge / 

 their muskets at the same instant, but are, nevertheless, heard in succession) s 

 does not seem to be in accordance with the phenomena. The passage of the 

 electric fluid through the air would be more aptly illustrated by a bow drawn \ 

 over the string of a violin, or the current of air driven by the mouth through a ) 

 wind instrument, or by a bellows through an organ-pipe. There would, ac- ( 

 cording to such analogy, be one sustained sound, instead of a succession or se- ) 

 ries of distinct sounds. It is true that, in the gravest note on aa organ, and eve* 

 in those produced on certain wind instruments (the trombone, for example), and 

 on the strings on the double base, the vibrations are distinguishable ; but these 

 vibrations do not seem to have any analogy to the series of sounds which form 

 the rolling of thunder. 



If this hypothesis, nevertheless, be admitted to explain the rolling of thunder, 

 the duration of the rolling will become an important element in determining the 

 minor limit of the space through which the lightning passes. Supposing that 

 no line drawn from the observer to the course of the lightnirtg is perpendicular 

 to it, it will follow that one extremity of the course is nearer than any other 

 point of it to the observer, and the other extremity more remote. The difler- 

 ence between the distance of these extreme points would be the length of the 

 flash, if its direction was immediately toward or from the observer ; and if it 

 have any other direction, this difference will be less than the length of the 

 flash. The duration of the roll of the thunder being the time sound would take 

 to move over the difference between the greatest and the least distance, this 

 difference may be computed, and thence a minor limit of the length of the flash 

 ^ may be obtained. 



From the observations of De L'Isle, it appears that the rolling of thunder, 

 ) observed by him in 17] 2, lasted in some instances forty-five seconds. Allow- 

 < ing eleven hundred feet for each second, this would amount to forty-nine thou- 

 ) sand five hundred feet, or very near ten miles. The length of the flash must, 

 I therefore, have exceeded this distance. 



I have, in these explanations, assumed that the loudest sound is that which 

 I proceeds from the nearest focus of sound to the observer. The loudness of a 

 S sound, however, depends partly on the temperature and hygrometric condition 

 J of the air at the place where the sound is developed. It might happen that ] 

 these conditions, varying in different parts of the air where the sounds are < 

 produced, would render more remote sounds sometimes louder than nearer 

 ones. 



One of the circumstances in the natural exhibition of lightning, which seems 

 not so satisfactorily explicable as most of the others, is the frequent repetition 



