Radiation emitted in Gaseous Explosions. 93 



of wave-length 2'S //. and 15 fi *), it will be clear from these 

 considerations that at any given temperature the rate of 

 partitioning will be slower the shorter the vibration period. 

 At any given tempera ture, therefore, the transfer to or from 

 the degrees of freedom corresponding to 2'S /n radiation will 

 be slower than that from the degrees of freedom corre- 

 sponding to 4*4 fi and 15 fi radiation — unless the temperature 

 is so high that the rate is very rapid for all types. 



17. Such a theory as that sketched in the preceding 

 sections affords a clear explanation of the reason for the 

 disappearance of the 2'S fj, radiation. At 1000° C. abs. 

 it may be supposed that the overwhelming proportion of 

 the molecular collisions are too soft to excite the type 

 of vibration corresponding to this radiation, while they 

 are sufficiently hard to excite the 4*4 /j, vibrations. As 

 the temperature increases, an increasing proportion of the 

 collisions becomes sufficiently hard to excite the 2'S ft 

 vibrations, and at 1200° C. abs. the volume of 2 '8 ft radiation 

 emitted (though still a small proportion of the total radiation 

 emitted) indicates that an appreciable proportion of the colli- 

 sions have become sufficiently hard to enable a transfer from 

 the translational and rotational degrees to go on fairly rapidly 

 so as to balance the loss by radiation. After this temperature 

 the ratio of the energy in the 2'S fi radiation to that in the 

 4*4 fi radiation indicates that the proportion of collisions 

 which are hard in respect to the 2'S fj, type of vibration 

 increases very rapidly as the temperature increases. 



18. Radiation due to chemical causes. — Assuming that 

 much of the energy of chemical combination passes in the- 

 first instance into the form of intra-molecular energy!, 

 there appears to be strong evidence in these experiments 

 that the distribution of energy from this cause over the 

 various vibratory degrees of freedom possessed by the 

 molecules is very much dependent upon the degree of 

 violence with which chemical combination takes place. 

 Were the distribution of energy over the vibratory degrees 

 of freedom independent of the violence of combustion, it 

 would have been expected that the ratio of the 2'S /jl radiation 



* The types of vibration corresponding to luminous radiation are not 

 dealt with' in this paper. The molecular collisions, even at 'the highest 

 temperatures developed in explosions, are too soft to excite these high- 

 frequency vibrations. 



t This includes both the rotational and the vibratory energy. There 

 is much experimental foundation for this hypothesis in respect to the 

 vibratory energy (see Phil. Mag. Feb. 1913, p. 263). 



