214 REPORTS ON THE STATE OF SCIENCE. 



the gas in this case was atmospheric, and the 30 cm. thickness in the 

 explosion vessel would be equivalent to perhaps 150 cm. of open flame 

 if absorption were simply proportional to density. According to Cal- 

 endar's experiment, such a thickness would be almost completely 

 opaque. It is possible that the later-al extension is sufficient to account 

 for this result. The open flame should be a cylindrical mass of dimen- 

 sions 150 cm. x 150 cm., instead of a long strip with a cross section 

 of 3 cm., in order to make the two cases strictly comparable. It will 

 be remembered that in the discussion above it appeared that the 

 laterally extended flame would seem to be more transparent. 



In the course of the year Mr. D. L. Chapman, of Oxford, and 

 Mr. H. E. Wimperis have joined the Committee. 



The Committee recommend that they be reappointed, and ask for a 

 grant of 1001. 



APPENDIX A. 



Radiation from Flames. By H. L. Callendar. 



In the course of my experiments in 1903-04 with a small petrol 

 motor of 2 - 36-inch bore on the variation of efficiency with speed, 

 I became convinced that the greater part of the loss of efficiency with 

 a small high-speed motor was practically independent of the speed. 

 Loss by radiation from the flame appeared to be one among the many 

 possible causes contributing to this result, and I accordingly made some 

 experiments on radiation from flames with a view to estimate the 

 pi*obable order of magnitude and the possible limits of the loss incurred. 

 The experiments were necessarily of a qualitative character, and could 

 not be directly applied to the calculation of the actual loss occurring 

 ' in an internal-combustion engine, but they appeared to indicate that the 

 effect was much larger than had generally been supposed, and could 

 not be neglected in a discussion of the heat loss occurring in a gaseous 

 explosion. 



Some of the results of these experiments were mentioned in the 

 discussion on a paper by Professor B. Hopkinson, ' Explosions of Coal 

 Gas and Air, ' 1 and a general summary was given in the discussion 

 on my paper, ' On the Effect of Size on the Thermal Efficiency of 

 Motors,' 2 from which the following is a quotation. 



' A large part of the energy of the flame during ignition exists in 

 the form of energy of vibration of the dissociated and recombining 

 ions, which is proved by the fact that a flame radiates energy more 

 intensely than a mass of inert gas at the same temperature. The 

 energy of vibration is realised as pressure, or energy of translation, only 

 in proportion as the ions combine and equilibrium is established. The 

 loss of thermal efficiency from this cause is merely another aspect of 

 dissociation or increase of apparent specific heat, and is not a loss of 

 heat at all, though it gives rise, as already explained, to a considerable 

 diminution of the thermal efficiency. But while the condition of flame 



1 Proc. R.S.A., 77. p. 400, April 1906. 

 3 Proc, Inst. Aut, Eng., April 1907, 



