ELECTRIC-SPARK PHOTOGRAPHS. 181 



four or five times as rapid as tlie motion of tlie Imllet forwards. A 

 new air wave is just beginning' to be created in front of the glass-coated 

 liead of tlie bullet and two highly inclined waves, one on either side of 

 the glass, reaching about three-quarters of the way to the edge, have 

 sprung into existence. These are more clearly seen in the next figure; 

 meanwhile it may be well to point out that the fragments of paper 

 wliich are following the bullets have in this case — as the card was 

 much nearer to the glass plate than in those previously taken — some 

 of them lost so much of their velocity and have in consequence lagged 

 behind in a still higher proportion than the others, that they are trav- 

 elling at less than 1,100 feet a second; the more backward ones carry 

 in consequence no air waves and there is no means of telling from the 

 photograph that they are moving at all. In PI. xii, fig. 1, the bullet 

 has struggled about half way through the plate. The waves on either 

 side of the plate have now reached the edge and are on their way back 

 towards the center again. They are caused in this way: When the 

 bullet strikes the plate the violent shock produces a ripple or tremor 

 in the glass which travels away radially in all directions, leaving the 

 glass quiet behind. The rate at which this ripple travels maybe found 

 from the angle which these new air waves make with the plate, for 

 taking any point on the plate and measuring up to the pcnjit where the 

 air wave meets the plate and also the distance in air to the nearest 

 point of the inclined air wave, we get two distances, the ratio of which 

 'is the ratio of the velocity of the disturbance in the glass to the veloc- 

 ity of sound in air. But much more tlian this is shown. An examina- 

 tion of the negatives or of a photographic print and even, but less 

 clearly, of the print in the text shows that these inclined air waves are 

 made up of a series of dark and light lines at a very slight iuclinatioii 

 to the air wave itself, so that as we travel along the air wave it is 

 alternately daik outside, and light outside. 



These indicate the successive positions in which the glass first moved 

 outwards to compress the air or first moved inwards to rarefy it so that 

 the wave length of the ripple nmy thus l)e found, and finally it is seen 

 that where the waves are waves of compression on one side of the plate 

 they are waves to rarefaction on the other, indicating that it was a 

 transverse and not a mere longitudinal disturbance that ran along the 

 plate from the center outwards and back again after reflection from the 

 edge. In addition to this fact that the reflected wave is still on its in- 

 ward course proves that up to this time the plate is whole, as a wave 

 can not be propagated in a broken plate. Plate xi, illustrates the 

 state of affairs when the bullet has traveled about 5 inches beyond the 

 plate. It has not yet emerged from the cloud of glass dust. The new 

 head wave is very conspicuous. In the original negative, about half- 

 way between the bullet and the plate, the inclined waves due to the 

 tremor in the glass plate may be detected, but tliey are too delicate to 

 be re-produced by the printing process. They supply the information as 



