322 
WiOFESSOE H. B. DIXON ON THE MOVEMENTS 
Tliese visible waves, accordiug to Oettixgen and Gerxet, are not a picture of the 
process of combustion itself, but are compression-waves moving through the products 
ot combustion after the explosion is completed. The explosion itself, they sav, is 
quite invisible. 
Fig. 5, Plate 10. is a repi-oductioii of their picture, showing the waves produced 
when the electrolytic gas is tired in the middle of a eudiometer 400 millims. long. 
The image of tlie spark is lengthened into a vertical line by retlexion from the 
sides of the eudiometer. Nothing else is visible on the plate for some time except 
a thin wavy line, which, according to Oettixgex and Gerxet, is due to the particles 
of salt I'aised to incandescence by the spark. Then after the lapse of second the 
material in the tube becomes luminous and a wave is seen starting from the upper 
end of the tube at d and traversing the gases to the lower end where it is reflected at c, 
and so on backwards and forwards, making nearly four comjjlete vibrations before it 
becomes non-luminous. The photograph shows very beautifully the passage of these 
compression-waves through the heated gases, and the way in which the mass of gas 
(carrying with it the luminous pai’ticles from the salt) follows the com^oression-waves 
backwards and forwards. But I can see no evidence in this photograph for the 
conclusion drawn by Oettixgex and Gerxet, viz., that a true detonation-w'ave has 
proceeded from the spark, and its compression-wave has traversed the tube several 
times befoi'e it becomes visible by raising the salts to incandescence. The tig. 5a 
(I'epi'oduced from their Memoir) shows in outline the course of one of these 
hypothetical waves starting downwards from the spark and being reflected at o, h, 
c, and d befoi'e it becomes visilile. According to this view, the wave has traversed 
the tube 3^ times before it is photographed, and the true chemical combustion gives 
no light. 
In the next photogi'aph reproduced (flg. 6) the gases were tired at a point 
a quarter of its length from the bottom of the tube. In this case there is a disjflace- 
ment of the wavy line joining the spark to the luminous jiortion, but hardly suflicient 
to justify the conclusion that the visible wave proceeding downwards fiom d is the 
residue of a detonation-wave starting upwards from the spai'k and being reflected at 
h, and again at c. But this is the interpi'etation put upon the photograph by 
Oettixgex and Gerxet, as is shown in tig. Ga. By constructing these invisible waves 
backwards from the visible ones, it is possible, according to Oettixgex and Gerxet, 
to arrive at the velocity of the true detonation-Avave at staiting. In this way they 
calculate that the explosion travels from the spark at a velocity of about 2550 metres 
per second, and criticise Berthelot’s statement that the flame increases regularly in 
velocity from the jioint of ignition. I will show in the sequel that the interpretation 
put upon these photogi'aphs by their authors is erroneous. The flame I'eally starts 
slovdy, but its rate of progress is remarkably aflected when it is reflected from the 
end of a tube. It is quite true that the flame of electrolytic gas when fii'st ignited 
has very slight luminosity, but this only holds-during the period of sIoav motion. The 
