342 
PROFESSOR H. B. DIXON OX THE MOVEMENTS 
Wliile our observations and those ot Le Chatelier are mainly in accord, vre 
difter from him in his conclusion that the wave of retonation (I’onde retrograde) is 
propagated in the hurnt gas; we believe it is propagated in the still hurning gas. 
With regard to the effect of collisions between two detonation-waves, Le Chateliee 
appears to consider the act of crossing of one wave by another to be sufficient to 
damp down their velocity ; we, on the other hand, regarding the “prolonged wave” 
of the one as the reflected wave of the other (which is a mere verbal distinction), 
and attribute the retardation of the reflected wave to its altered chamcter and to the 
movement of the gas which meets it. 
PAET VI. 
(4n the Initial Phases of the Explosion. 
. {In conjunction with P. H. Jones and J. Bower.) 
The interest attaching to the development of the explosion, as well as a desire to 
investigate the anomalies shown by some of our pictures, led us to attempt to 
photograph the flame from the beginiiiug. 
In fig. 28 (Plate 13), the initial flame, starting from the left-hand lower corner of 
the photograph, is overtaken by a faster-moving flame. Fig. 39 (Plate 14) shows a 
similar phenomenon in the right-hand corner. On photographing in a straight glass 
tube the region of exjJosion prior to tlie detonation, we found (fig. 52, Plate 16) the 
initial flame was overtaken by a bright and weU-marked faster flame, and, at their 
point of meeting, a reflected wave was driven back, and the advancing flame of 
explosion became faster and more luminous. What is the origin of this faster wave 
which overtakes the advancino; flame ? 
A mixture of C 3 mnogen with twice its Amhune of oxvgen was found to give a 
sufficiently luminous flame from its start to be photographed on the moving film, but 
mixtures of carbon disidphide Avith oxygen gaAm still better images. Fig. 53 
(Plate 16) Avith the cyanogen mixture, and fig. 54 Avith the carbon disulphide mixture, 
shoAved that the bright Avave which overtook the initial flame came from the end of 
the tube near the firing wires. The wires in the tube used Avere sealed through the 
glass 3 inches from one end, and the glass had been so much distorted at the point of 
sealing that a dark band aars shoAAm on the photograph Avhere the Avires were 
inserted. 
Another tube Avas prepared with firing Avires inserted 4 inches from one end; the 
slight distortion at the AAures acted like a convex lens and increased the light at this 
point. Fig. 55 shows the development of the explosion Avith CSo fl- 5 O 2 ; fig. 56 
