308 



F. J. Rogers — Magnesium as a Source of Light. 



radiation for the different wave lengths of the luminous and 

 heat spectrum, found the radiant efficiency of gaslight from an 

 Argand burner to be - 024. 



Considerable difficulty was experienced in measuring this 

 ratio for magnesium light on account of the great variability 

 of the magnesium flame. The galvanometer needle instead of 

 moving steadily or by steps to a maximum and then remaining 

 there, was almost constantly in motion. Furthermore the 

 radiation with and without the alum cells had to be measured 

 successively, and there was no certainty that the condition of 

 the flame was the same in the two cases. The best that could 

 be done was to take a large number of observations and dis- 

 card a few that were very far from the mean. 



Table IV. 



No. of 

 Experiment, 



No. 1 



« 2 



Galvanometer deflections. 

 Two alum Direct radia- 



cells tion R. of cir- 



R. of circuit cuit=8030. 



=803. 



67-4 



n-5 



76"5 



83-5 



Light trans- 

 mitted through 

 alum cells divi- 

 ded by total radia- 

 tion. 



•088 

 •093 



Preceding 

 ratio correc- 

 ted for heat 

 transmitted. 



•085 

 •090 



No. 



Same as above but icith diaphragm. 

 16-5 17-9 -092 -089 



21-7 22-7 -095 -092 



Radiant effi- 

 ciency (preced- 

 ing ratio corrected 

 for light reflected 

 and absorbed.) 



•133 

 •140 



•139 

 •144 



No. 



Same as above but icith galvanometer JVb. 2. 



R. of circuit=37-46 R. of circuit 147-46 



5 18-0 48-6 -094 '091 -142 



" 6 17-8 49-1 -092 -089 -139 



" 7 12-4 40-8 -077 '074 -115 



8 14-1 41-2 '087 '084 -131 



Same as above, sensitiveness of galvanometer changed. 



R. of circuit=20-46 R. of circuit 207' 46 



No. 9 31-0 40-5 -076 -073 '114 



Average by method of permanent deflections, '133 



In Table IV, are given nine determinations of the radiant 

 efficiency of magnesium light. In the first two no diaphragm 

 was used, but in every other case a diaphragm of 10 X 15 mm 

 opening was placed in front of the ribbon of burning magne- 

 sium. This reduced the fluctuations caused by the varying 

 length of the incandescent strip of magnesia.* 



It has been provedf that when a galvanometer and thermo- 

 pile are used together the " first swing " of the galvanometer 



* It is perhaps a misnomer to call this strip of incandescent magnesia a " flame." 

 Neither the combustible nor the product of combustion is a gas and the burning- 

 magnesium unlike a flame hangs downward. 



f Ernest ilerritt, note on galvanometer used with thermopile, this Journal, xli, 

 p. 417. 



