34(3 
rEOFESSOE II. B. DIXOX OX THE MOVEMEXTS 
centre of a tube. If the flame overtakes the sound-wave before it reaches the ends 
of the tube, as in fig. 61, the explosion exhibits intense retonation-waves followed bv 
i-apid cooling; but when the sound-waves are first reflected from the ends of the 
tube, as in fig. 65, the explosion shows less intense waves and a longer illumination.''^ 
Tliese differences in the mode of burning must affect the maximum temperatures and 
pressures produced in an explosion. 
Gases Fired at Open End of Tube. 
We have made a few experiments on the appearances presented when gaseous 
mixtures are fired at or ]iear the open end of the tube. 
As is well known, from Le Chatelier’s researches on the vibratory period, a 
juixture of carbon disulphide and nitric oxide gives oscillations of large amplitude. 
We show one photograph of the explosion of the mixture OS., -j- 8NO fired at the 
open end of a tube 5 feet long and 1 inch bore. Only the last foot of the tube is 
shown in fig. 68. 
The effect of firing the mixture CoN^ -f- 0,, 3 inches from an open end is the same 
as that of firing it at the closed end. Fig. 69 shows the detonation set up in about 
8 inches, and of course no rebound from the open end. The mixture 2CoHo + 30,, 
lighted at the open end produced the detonation in 4^ inches; lighted by a spark 
3 inches from the ojjen end, it set up the detonation in 2^ inches from the spark. 
PART Vll. 
Further Experiments on the Initial Phases. 
[In conjunction with B. DxiwsoN, B.Sc., and L. Bradshaw, B.Sc.) 
]. Le Chatelier’s Hypothesis of Discontinuity in the Explosion. 
]\Iany of the photographs previously referred to show a peculiai'ity at the point 
where a less luminous line is succeeded abruptly by a more luminous one. The lines 
]diotographed do not appear continuous, but the moi'e luminous line appears to start 
from a point not yet reached by the less luminous one. Figs. 28 to 31 (Plate 13) 
illustrate this discontinuity. The point of collision also of two waves appears to 
pi'oject in front of the waves which are meeting. Figs. 29 to 31 illustrate such a 
collision. It appeared to me at first as if these appearances might lie due to invisible 
waves advanciiip’ in fi'ont of tlie visible ones, but as I found that thev oidv showed 
* The very interesting jhiotograplis, obtained l.)y Mr. Petavel, of tlie movements of a piston connected 
vith a chamber in vhich gases under high in'essure are exploded, sliow vibrations w Inch may be due to 
reilexiou-waves. 
