the Occurrence of Enhanced Lines in the Arc. 869 



Bunsen cone o£ such characteristic arc lines as the magnesium 

 triplets \ 3838 and X 5184, and their absence from the hottest 

 part in the mantle of the same flame, are results which go far 

 to emphasize the necessity for exercising the greatest caution 

 in selecting spectrum lines for the purpose of establishing a 

 temperature classification of the stars'. The relative sensitive- 

 ness of the various types of magnesium lines to electric and 

 chemical actions is brought out in the following table (p. 870), 

 which has been constructed with the help of data obtained in 

 the course of the present research and of those resulting from 

 observations by Messrs. Hartley and Hamage *■, de Watte- 

 ville |, King %, and Hemsalech §. 



It will be seen from this table that the only types of 

 magnesium lines emitted in the absence of electric or special 

 chemical actions are the flame line A 4571 and the series 

 triplets. But whereas the former is apparently not affected 

 by electric actions, the triplets, on the contrary, show marked 

 increase in brightness as the intensity of electric actions rises. 

 Also the special chemical actions prevailing in the inner cone 

 of the air-coal gas flame bring the triplets into prominence 

 already at & temperature which by itself would be unable to 

 excite them. Hence, if in a source of light the triplets are 

 observed to be very intense as compared with the line X4571, 

 we must suspect the presence of chemical or electric actions 

 in that source. The electric actions, as has been shown by 

 one of us ||, may be relatively feeble, provided the metal 

 vapour be at a high temperature. If, however, the source 

 of light emits in addition to these two types of lines also 

 those of the Kydberg series and the spark line, then, as our 

 table indicates, the electric actions must be of a much more 

 vigorous kind than in the first case. Should the spark line 

 A- 4481 appear symmetrically broadened and the lines of the 

 Bydberg series winged on the red side, it is most probably 

 an indication that an electric current flows through the source 

 or part of it at a high potential gradient under the action of 

 a powerful electric field. The temperature of the vapour in 

 the source may in this case be relatively low. If, on the other 

 hand, A, 4481 be a sharp narrow line, relatively feeble as'com- 

 pared with the triplets and the lines of the Rydberg series, 

 and if, further, the lines of the latter be fairly well defined, it 



* W. N. Hartley and H. Rainage, Trans. Royal Dublin Soc. ser. 2, 

 vol.vii. p. 843 (1898-1901). 



t C. de Watteville, Phil. Trans. Roy. Soc. A. vol. 204. p. 151 (1904). 

 X A. S. King, Astrophvsical Journal, vol. xlviii. p. 79 (1918). 

 § G. A. Hemsalech, Phil. Mag. vol. xl. p. 307 (1920). 

 Ij G. A. Hemsalech, loc. cit. 



