OF THE FLAME IX THE EXPLOSION OF GASES. 
321 
To make clear the movements studied by Mallard and Le Ch atelier I reproduce 
on Plate 10 four of theu’ photographs. 
In fig. 1 the mixture of carbon disulphide with six times its volume of nitric oxide 
is ignited at the open end of a tube 3 metres long and 20 millims. in diameter. The 
tube is made up of three pieces, of 1 metre each, fastened together by rubber rings, 
which, eclipsing the flame as it passes, show as blank bands on the photographs. 
The flame advances with a uniform velocity of 1’25 metres per second to the point b, 
where the rate increases and vibrations begin. From this point the rest of the 
combustion takes place with strong vdbrations of the flame. 
In a tube of 10 millims. diameter the jieriod of unifoim velocity is shorter (fig. 2). 
The vibrations die down and recommence, but after the flame has traversed a little 
over 1 metre the flame is extinguished. 
o 
Fig. 4 is the photograph of the ex|)losion of carbon disulphide and oxygen fired at 
the open end of the tube. A short j^eriod of uniform jirogression is followed abruptly 
by a flame which appears to be propagated instantaneously. 
Fig. 3 shows the effect of igniting the mixture of carbon disulphide and nitric 
oxide near the closed end of a tube 2 metres in length. The velocity of the flame 
continually increases from the jioint of inflammation, a. This photograph"’' clearly 
shows a “ rebound ” wave, c d, from the end of the tidie, which was not broken like 
the one described in the text (‘llecherches,’ p. fl). 
Experiments of A., vox Oettingex and A. vox Gerxet. 
Oettixgex and Gerxet t set out to prove tlie truth of Buxsex’s principle ot 
successive jJai'tial explosions. This discontinuous step-like condjustion (“ welches wir 
juit neuen Methoden experimentell zu prlifen unternahmen ”) should yield evidence 
of its real existence if the flame of the exjilosion were analysed by a lotating mirror. 
By an ingenious arrangement they contrived to pass a spark through a eudiometer 
tube at the moment when the image of the tidie was thrown by the lotating mirror 
into a camera, so tliat the light of the flame might be drawn out and its movements 
recorded on a photographic plate. But altliough the flame of electrolytic gas 
ajjpeared intensely bright, its spectrum only gave the sodium and calcium lines, and 
the most sensitive photographic plates showed “ hardly a trace of the process. ’ 
Failing to 2 )hotogTaph tlie flame itself, they added finely divided salts to the tube, 
and found that the most brilliant pictures were given by cuprous chloride. 
The pictures show the passage of waves sharply reflected liackwards and forwards 
from the ends of the tube, and gradually diminishing in intensity and velocity. 
* This photograph is not descriljed in the text of MallAKD and Le Cii.ITELIEr’s book. 
M. Le Chatelier writes to me that my interpretation of the photogra23h is correct, I did not notice 
this “ rebound” wave in the photogra^A until after my experiments of 189G. 
t “Feber Knallgascxplosion,” ‘ Annalcn der Phys. und Chemie,’ 3-3, 586 (1888). 
VOL. CC.—A. 2 T 
