:276 



Mr. Dugald Clerk 



[Jan. 29, 



■flame its rate of movement continually increased until the whole of 

 the vessel was filled. Had combustion been complete at this stage, 

 then no further rise of pressure could have occurred. It was evident, 

 however, that although every part of the vessel was filled with flame, 

 t-he chemical combination had not ceased. The flame, at first very 

 rapid in its rate of combination, reduced as combination proceeded, 

 because of the increasing presence of carbon dioxide and steam-form- 

 ing diluents to the inflammable gases still remaining to be combined 

 with the excess of oxygen. It showed, indeed, that although a gaseous 

 •explosion is quick, yet chemical combination, even under such 

 circumstances, slows down at high temperatures just as it does in 

 other chemical combinations occuring at low temperatures. Although 



Fig. 6. — Explosion and Cooling Curve of Gas and Air Mix- 

 ture (1 Gas and 9 Air) at Atmospheric Pressure, with 

 temperatures reached by platinum resistance thermometers. 

 (Hopkinson.) 



Clerk thus inferred many years ago that at a certain point on the 

 rising line the flame had completely filled the vessel, yet this was not 

 definitely proved until Hopkinson made his experiments, which will 

 now be described. 



Hopkinson's large explosion vessel is shown at Fig. 7. It was, 

 -as has been stated, 23-4 inches diameter and 28*75 long. Electric 

 ignition was effected by a spark passed at the point A in the centre 

 of the vessel. Three platinum wire resistance thermometers were 

 ■arranged respectively at B, C, and D— B close to the spark, C 10 



