Manchester Memoirs, Vol. xlii. (1898), No. 7. 3 



to one another, fastened together by lead bends and caps 

 which fitted over the glass, the mixture being fired in the 

 uppermost tube. 



What we required was a junction of lead pipe to glass 

 tubing, both the lead and glass having an internal bore of 

 14^ in., but having different external diameters — the lead 

 ^in., and the glass ^in. The mode of joining in these 

 cases where a stoppage occurred was to take a short piece 

 of lead piping, which just slipped over the glass tubing, 

 and to beat down one end until the bore at that end was 

 only }(\n. ; this was then soldered to the lead piping of 

 i^in. internal diameter, a short piece of rubber tubing was 

 drawn over the end of the glass tubing, and this was finally 

 thrust into the lead cap. These junctions were not quite 

 rigid, and the ends of the tubes not exactly flush. 



The pictures given show clearly the partial stoppage 

 of the explosion wave at such a joint, the point of redeter- 

 mination being indicated by a definite and increased 

 speed of combustion, represented by the straight and 

 intensely luminous line as distinct from the curved and 

 less luminous line of the recouping period. 



Further, a new and very luminous wave is observed 

 starting from the point of redetermination of the explo- 

 sion wave proper, and travelling in the opposite direction 

 with a speed almost, if not quite, equal to that of the 

 explosion wave itself 



The point, however, bearing most directly on Dixon 

 and Cain's paper is this — that tubes which stood the pres- 

 sure of the explosion wave itself, almost invariably broke 

 at this point of redetermination. 



Thus, in their experiments, they might really be 

 measuring, not the pressure of the explosion vv-ave, but 

 this greater pressure always associated with the re-institu- 

 tion of the wave after a partial stoppage. The figures 

 obtained forced them to the conclusion that there was no 



