378 Messrs. W. E. Wilson and G. F. Fitzgerald. 



boiling carbon. The most modern determinations give this tem- 

 perature of the crater as about 3300 3500 C.* 



If this is the true boiling point of carbon, it is then clear that 

 solar physicists must find some other substance than solid carbon 

 particles to form the photospheric clouds in the sun, as the tempera- 

 ture of this layer is most probably not below 8000 C.,f unless, indeed, 

 the pressure in the solar atmosphere is sufficient to raise the boiling 

 point of carbon to about this temperature (see p. 381). It is in 

 order to throw some light on this subject that these experiments 

 were undertaken. 



The gas used in our first experiments was nitrogen, and we found 

 that the radiation from the crater fell off in a. most remarkable 

 manner whenever the pressure was raised in the box surrounding 

 the arc. This falling off was not due to any very large extent to 

 visible cloud or smoke, and the crater seemed so much reduced in 

 temperature as to glow with only a red heat. This seemed to show 

 that the temperature of the crater depends on how much it is cooled 

 by the surrounding gas, and not on its being the temperature at 

 which the vapour of carbon has the same pressure as the surrounding 

 atmosphere. 



It was found that we were limited to pressures not exceeding about 

 20 atmos., as at this pressure we could not withdraw the negative 

 carbon sufficiently to see into the crater without the arc breaking. 

 We were then only able to obtain a current from a battery of accu- 

 mulators which had an E.M.F. of 110 volts. Since then we obtained 

 a Crompton dynamo which could give 300 volts and 15 amperes, and 

 which was driven by a turbine. 



From the great difficulty of obtaining a sufficient quantity of pure 

 nitrogen under pressure, we obtained a 20 ft. cylinder of air com- 

 pressed to 120 atmos. With this we tried a series of experiments, 

 and these at first seemed to corroborate our former ones, in which we 

 used nitrogen, but we found that at any rate some of the radiation, 

 and possibly a great deal of it, was cut off by the formation of what 

 appeared to be red fumes of N"O 2 . We found no absorption from 

 this cause so long as the pressure was nearly atmospheric, but at 

 about 100 Ibs. pressure this gas was formed with great rapidity, and 

 undoubtedly cut off a great deal of the radiation. We easily con- 

 firmed our belief in the presence of this gas by its well known 

 absorption spectrum. 



Lest heat dissociation might cause an apparent increase in the 

 amount of NO 2 , we tried heating some of this gas in a flask. We 

 observed that when hot the brown fumes became golden yellow, and 



* Wilson and Gray, ' Koy. Soc. Proc.,' vol. 58 j Violle, ' Journ. de Phys.,' 3rd 

 series, vol. 2, 1893, p. 545. 



f Wilson and Gray, < Phil. Trans.,' A, vol. 185, 1894. 



