the Theory of Luminous Flames. 101 



intensity of light in experiments V. and VI. with those in the 

 preceding experiments, it is seen that for equal consumption 

 of gas the intensity of light is much smaller in the former than 

 in the latter cases. The outer surface of the flame is too 

 small in proportion to the total volume of gas to allow of such 

 energetic action as is required for the production of a tempe- 

 rature sufficiently high for the maintenance of great lumino- 

 sity ; but inasmuch as a large volume of gas is present, artifi- 

 cial raising of the temperature of that gas very greatly increases 

 the luminosity of the flame: this effect, however, must be 

 decreased by increasing the velocity of the issuing gas ; expe- 

 riment V. shows that this is really the case. As the velocity 

 of outflow increases, the increase in luminosity becomes much 

 less in proportion to what it was at the beginning of the ex- 

 periment, until a point is reached, after which the luminosity 

 begins to decrease, the flame finally burning blue. If the gas 

 be passed through a cold tube, this decreased luminosity accom- 

 panying increased velocity of the issuing gas becomes much 

 more marked ; hence after a certain point the effect of increased 

 gas-consumption with a hot tube in increasing luminosity must 

 again surpass that with a cold. This is seen by throwing the 

 results of the Table into graphic form and comparing the 

 curves representing the increase in luminosity for equal but 

 increasing gas-consumption according as hot or cold tubes are 

 employed. At first the curve of the hot tube keeps far ahead 

 of that of the cold ; as the consumption of gas increases, the 

 curves become equal (when 40 or 45 litres of gas are con- 

 sumed); but after this point the hot-tube curve again passes 

 ahead of that of the cold tube. 



Experiments I., II., and III. show that heating the tube 

 through which the gas passes causes a decrease in the quan- 

 tity of gas consumed. The decrease is more marked in the 

 case of narrow than of wide tubes. 



These experiments show that by considerably raising the 

 temperature of the issuing gas and of the burner, a marked 

 increase in the luminosity of the flame becomes apparent. 

 The increase of temperature was too small in the experi- 

 ments of the English Commission to allow of the effect being 

 noticed. 



I have also made experiments with flat flames, produced by 

 allowing the gas to issue from a narrow platinum tube the 

 end of which was hammered together so as to leave only a 

 narrow slit. The following Table contains the results ;— - 



