864 



SCIENCE 



[N. S. Vol. XXX. No. 781 



given out is not so great as is often as- 

 sumed. It does not give effects at all com- 

 parable with full sunshine. While doubt- 

 less the intrinsic brilliancy is very high the 

 duration of the flash is small, generally 

 only a minute fraction of a second. In 

 photographs of lightning the landscape is 

 generally seen only in outline or poorly 

 lighted by the discharge. In the daytime, 

 when the clouds are not dense enough to 

 greatly darken the sky, the flash loses most 

 of the blinding character it has when seen 

 in the blackness of night. Similarly, the 

 sound of thunder, though of terrifying 

 quality, is not extraordinarily loud. It is 

 a common experience when traveling in a 

 train to note that the sound of even near-by 

 flashes is smothered by the roar of the 

 train so that no thunder is heard. The 

 noise of thunder can not be due in any 

 part, as is sometimes erroneously assumed, 

 to collapse of the air upon itself and into 

 a partial vacuum left by the spark. I have 

 seen this error even recently repeated and 

 even extended to include all the noise of 

 thunder as due to such collapse. "When, 

 however, we consider that in a minute frac- 

 tion of a second the air in the path of the 

 discharge is so highly heated that, if it 

 were confined, its pressure due to heat ex- 

 pansion alone would rise to more than ten 

 atmospheres we can readily understand the 

 explosive shock given to the surrounding 

 air and the propagation therethrough of an 

 intense air wave. In fact such waves from 

 electric spark discharges and from dyna- 

 mite explosions have been clearly recorded 

 by photography. Moreover, that the col- 

 lapse of the air after expansion can have 

 little or no effect in the sound production, 

 follows from the fact that the heated gas 

 streak left in the path of the discharge 

 takes an appreciable time to cool on ac- 

 count of its low radiating power. This is 

 shown by the observation that a lightning 



discharge in dusty air is often succeeded 

 by a luminosity of the streak which per- 

 sists for a perceptible time and slowly 

 fades away like the luminous trail of a 

 meteor. 



Another common misconception is that 

 the prolonged rolling character of thunder 

 is due to reverberations or echoes. In 

 mountain regions with steep rock walls 

 such reverberations possibly contribute to 

 the effect, but it is now clearly recognized 

 that a sufficient single explanation suffices 

 for most cases. Owing to the great length 

 of the lightning spark or path, we receive 

 the sound from the nearer parts of the dis- 

 charge far in advance of that from the 

 more remote portions, and between these 

 sounds are those from parts of the path at 

 intermediate distances from the observer. 

 It follows from this that no two observers 

 at a distance from each other hear the 

 same succession of sounds in the thunder 

 of a discharge. "Whenever portions of the 

 discharge path are situated or extended in 

 an approximate direction at right angles 

 to the line from the observer, the sound 

 from that part of the path is louder or of 

 high amplitude owing to the sound from 

 that part of the path reaching the observ- 

 er's ear at the same instant. "Whenever the 

 path leads directly away from the observer 

 the amplitude is less, the sound is less ex- 

 plosive and takes the character of an ex- 

 tended roll or rumble. 



It will be seen from this that every twist 

 and turn and every change of direction of 

 the spark path with respect to the observ- 

 er's position gives a varying loudness and 

 sequence of sounds. Every branch of the 

 main discharge in like manner records its 

 position and direction, its twistings and 

 bendings in these sound vibrations and 

 sequences. It would seem possible even to 

 record on a phonograph noises from sparks 

 invisible to the eye and map the positions 



