of Light Radiations emitted by Iron Vapour. 9 



discharge circuit, and in series with the spark gap, is a self- 

 induction coil of 0'05 henry. The condenser receives its 

 charge from the secondary of a 10-in. Apps coil, the primary 

 of which is fed with pulses of continuous current at the rate 

 of about 5 per second through the medium of a Rochefort 

 plunger break, working in an atmosphere of coal-gas. / 



Reduction of spectrograms. 

 The plates were measured on a Perreaux dividing engine* 

 which had been fitted with a travelling platform for the 

 purpose. Since it was merely a question of identifying 

 known lines, the wave-lengths were derived by linear inter- 

 polation. The plan adopted was similar to that laid down in 

 my thesis (page 56). 



Relative intensities of lines. 

 These, of course, refer to the photographic intensities, and 

 are estimated on quite an arbitrary scale. Feeble lines are 

 marked ^, ver}' feeble ones and 00. Lines of moderate 

 strength 1, 2, 3, and 4 ; strong lines 6, 8, and 10 ; extra 

 intense 12 and 15. 



3. General Character of the Spectrum emitted by 

 Iron Vapour in the Air-coal Gas flame. 



Difference between cone and ft am e lines. 

 De TVatteville's observations had brought out the inter- 

 esting fact, already mentioned, that the majority of lines are 

 emitted by the region of the blue cone, so that, when an 

 image of the flame is formed on the slit parallel to the latter, 

 the cone lines are seen to reach only as far up as the bands 

 of the Swan spectrum. The flame lines, on the other hand, 

 pass beyond the boundary of the cone region into the flame 

 above. The flame lines are alone observed when the sub- 

 stance to be examined is introduced into the flame by any of 

 the older methods — for example, by means of a platinum 

 wire. 



Flame lines emitted by cone. 

 Since the cone of an ordinary flame is always and every- 

 where surrounded by a layer of flame, we always see the 

 flame lines superimposed upon the cone lines, and it is 

 impossible in these circumstances to decide whether flame 

 lines are likewise excited in the cone or whether the latter 

 emits solely its characteristic lines. In order to settle this 



* A description of the instrument is given in Stewart and Gee's 

 'Practical Physics,' vol. i. p. 16 (1885). 



