OF THE FLAME IN THE EXPLOSION OF GASES. 
825 
two volumes of oxygen travelling from right to left at the bottom of each 
photograph. In fig. 7 the wave strikes the end of the tube (a metal stopper being 
used), and a reflected wave is thrown back. The dark Imnd was 35 centims. from the 
stopper. Fig. 8 shows the reflected wave between 35 and 70 centims. from the 
closed end, and fig. 9 shows the reflected wave travelling between marks placed at 
70 and 105 centims. from the closed end. The film was moving slower when fig. 7 
was photographed than in the other two cases. 
The jdrotograjDhs in figs. 10, II, 12, and 13 show the reflected wave (in different 
mixtures of gases) travelling Ijackwards (from left to right) between 35 and 70 
centims. from the closed end. When the end of the tube is open only a faint flame 
is projected from the tube, and no bright reflected wave is seen (fig. 14). A cork, 
- loosely fitted, is sufficient to send back a Inight reflected wave (fig. 15). Even when 
the tube is fractured by the explosion, the detonation-wave can be photographed and 
an indistinct reflected wave is visilde (fig. IG). 
The first points noticed in the photographs were (l) the sharpness witli wliicli the 
luminosity is set up, and (2) the uniformity of the detonation-wave. There is no 
evidence of any gradual heating up of the gases, l)ut on the contrary the temperature 
appears to spring to its maximum with aljrupt suddenness. This is, of co\irse, in 
accordance with the views published by Berthelot and by myself as to the character 
of the detonation-wave, which we believe to l)e jjropagated l)y the shock of the 
molecules themselves moving forward with the velocity due to the whole heat of the 
chemical combination. The gas ignited by the detonation-wave (including dust and 
particles knocked off the tidies) remains luminous for some time after the wave has 
passed. As had been shown in the ‘‘ window ” experiments (previously referred to), 
cyanogen burning to caihonic acid left a longer trail of liglit than Avlien Vmrut to 
carbonic oxide only, and the most prolonged images were o])tained witli the two 
mixtures CoN,j + 20^ and CSo -f 3(4^. 
Many of the photographs show very distinctly the movements of the gas en masse 
as it follows up the detonation-wave, comes to rest, and swings back again. Fig. 17 
(taken in a long tube) shows tlie movements of the gas undisturhed l)y any reflected 
wave. These movements are analogous to the forward and backward movements in 
air produced by a vibrating Ijody. According to the kinetic theory, the mean motions 
of the molecnles of a gas may l)e resolved in any direction into equal and opposite 
movements. When a compressed tuning-fork is released, the forward movements of 
the molecules in contact with the piong have added to them the motion of the fork, 
and by exchange of velocities this added velocity is propagated from molecule to 
molecule, each molecule swinging forward with the increased velocity and returning 
with its normal velocity. In a sound-wave a numher of these forwaitl impulses is 
imparted to the molecules as the prong moves forward, and therefore the molecules 
move forward, causing a compression-wave, and afterwards, as the prong moves 
backwards, the reverse effect is produced, and a rarefaction follows the compression- 
