EXPLOSIONS OF COAL-GAS AND AIK. 407 



The accuracy of the figures in this table is not insisted upon, since they depend 

 upon K and It, which have been calculated from the observed values of the intrinsic 

 radiance from two lengths of gas only (except in the case of the mixture at 

 atmospheric density). I think, however, they definitely show that were the gaseous 

 mixtures perfectly transparent the intrinsic radiance from a thickness inversely 

 proportional to the density would decrease as the density increases, at any rate 

 within the limits of density in these experiments. The increasing emission from 

 30/D cm. of the gaseous mixture as D is increased (see chain-dotted curves, fig. 16) 

 must, therefore, be wholly due to the increasing transparency of 30/D cm. as D is 

 increased (cf. Tables XVII. and XVIII.). It should be noted that the equation 

 found above, connecting the intrinsic radiance with the density, viz., It = &D 025 , holds 

 only for 30/D cms. From small lengths of 1/D or 2/D cm. the intrinsic radiance 

 decreases as D increases. 



SUMMARY OF RESULTS AND SHORT THEORETICAL DISCUSSION. 

 The following are the main results obtained from these experiments : 



Part I. When mixtures of coal-gas and air of various strengths at atmospheric 

 density are exploded in the vessel when its walls are blackened over with a thin 

 layer of dull black paint 



(i.) The total amount of heat lost by radiation to the walls of the vessel up to the 

 moment of maximum pressure is roughly proportional to the product of the third 

 power of the maximum absolute temperature attained into the " time of explosion." 



(ii.) The total radiation lost to the walls during explosion and subsequent cooling 

 is about 25 per cent, of the heat of combustion of the gas present in the vessel. 



(iii.) The emission of radiation in the initial stages of cooling after explosion is a 

 function of the time from ignition as well as of the temperature. The emission varies 

 very rapidly with the temperature and the time from ignition. 



(iv.) In weak mixtures (and probably also in strong mixtures) the rate at which 

 radiation is emitted is a maximum some time before the attainment of maximum 

 pressure, and probably occurs at the moment when the flame fills the vessel. 



(v.) Weak mixtures radiate much more powerfully in the initial stages of cooling 

 after explosion than stronger mixtures do when they have cooled to the same 

 temperatures as the weaker mixtures have in this epoch. 



(vi.) CO 3 emits radiation about twice as strongly as an equal volume of water 

 vapour at the same temperature does. 



In explosions of mixtures of the same strength but of various densities 



(vii.) The total heat lost by radiation per cent, of the heat of combustion of the gas 

 present in the vessel up to the moment of maximum pressure decreases as the density 



mrivasrs.' 



