481 . REPORTS ON THE STATE OF SCIENCE. 



the increase of the specific heats of steam and carbon dioxide with 

 temperature and the small if not negligible influence of dissociation. 

 They calculated that the temperature of the flame when moist electro- 

 lytic gas is exploded in a closed vessel at atmospheric pressure is about 

 3350°, and that the mean molecular heat of steam at constant volume 

 between 0° and 3350° is 16-6. In the case of the mixture 2CO + 2 

 they concluded that the molecular heat (C a ) of carbon dioxide rises 

 to 13'6 at 2000°, above which temperature dissociation comes into 

 play. 



The experience of all subsequent investigators has confirmed that 

 of Bunsen and the French savants, in so far as the facts of the case 

 are concerned. It may be taken as commonly agreed that the maximum 

 effective pressures recorded when gaseous mixtures are fired in closed 

 vessels are always considerably less than those calculated on the 

 assumption that the whole heat of combustion is communicated without 

 loss to the products, and that the specific heals of the products do not 

 vary with temperature. Thus in the case of Mr. Dugald Clerk's 

 experiments, where hydrogen and air mixtures were exploded at 

 atmospheric pressure in a closed cylindrical vessel 21 cm. long, 

 17'75 cm. diameter, and 5 - 2 litres capacity, the maximum pressures, 

 recorded by a Richards' indicator making a graph on a revolving drum, 

 varied between about 50 and 60 per cent, of those calculated on the 

 above assumptions, as follows: — 



Mr. Dugald Clerk's Experiments (1884-85). 



Vols. Air to 1 Vol. Hydrogen .. 

 2£ 4 6 



Maximum pressure found . . 6-44 . . 5-63 . . 3-80 I Atino- 

 Maximum pressure calculated on spheres 



the above assumptions . . 13 - . . 9 - 45 . . 70 



Another feature (since confirmed by many subsequent observers) 

 brought out by Mr. Clerk's experiments was the very short time required 

 for the attainment of maximum pressure relative to the subsequent 

 cooling period. Thus in the case of the mixture of 1 vol. hydrogen with 

 4 vols, air, the maximum pressure of G8 lb. per square inch above the 

 atmospheric was attained in - 026 sec, whereas the subsequent cooling 

 period occupied l - 05 sec, or forty times as long. 



The great disparity between the found and ' calculated ' maximum 

 pressures lias been attributed by the various investigators concerned to 

 one or other of the following causes, namely: — 



1. To the marked increase in the specific heals of steam and carbon 

 dioxide with temperature. That this is a true cause is now generally 

 admitted ; as, however, the subject was fully dealt with in the first report 

 of the Committee appointed by the Association in 1907 for the 

 investigation of gaseous explosions, it need not be considered in any 

 detail here. 



2. To the fact that in ordinary gaseous explosions, where detonation 

 has not been determined, combustion is by no means instantaneous, and 

 may not be completed within the period required for the attainment of 



