POLAH LINE.S IX ARC SPEOTUA. Ill 



Besides the theoretical imjjortance of solar study in its relation 

 to the problem of stellar and inorganic evolution, there is the reason- 

 able hope that a knowledge of the relationship between solar and 

 terrestrial phenomena may prove of practical service to mankind. It 

 is interesting to note that India has erected a solar observatory, in 

 the belief that it will ultimately prove of value in famine prediction. 



10.— POLAR LINES IX ARC SPECTRA. 

 Bii W. GEOFFREY DUFFIELD, D.Sc. 



Several writers have chronicled the occurrence of " spark " lines 

 in arc spectra. Fowler* has described the appearance in the spectrum 

 of an iron arc of lines which are strongest at the poles, and diminish 

 in intensity as they approach the centre, being, however, stronger 

 on the positive than on the negative pole. He investigated the region 

 F to C, and pointed out the identity of these lines with the enhanced 

 lines of iron and with those lines that are weakened in sun spots. 

 The ultra-violet region of the spectrum of the iron arc has been investi- 

 gated by the writer,t a vertical image being focussed upon the 

 vertical slit of the 21-^-ft. Kowland grating spectograph in the Man- 

 chester University, the length of the arc being so adjusted that the 

 tip of each pole was just included upon the slit. The astigmatism 

 of the grating was not sufficient to mask the phenomenon. Many 

 lines appear only on the tips of the poles in these photographs, but 

 they differ from Fowler's in tliat tliey are of nearly the same intensity 

 on the two poles. The arc was supplied with current from the cor- 

 poration mains at 110 volts ; this was continuous, but, to ensure that a 

 superinduced alternating current was not disturbing the continuous 

 current and producing a weak spax-k discharge, several photographs 

 were taken when the arc was run from the storage batteries, and the 

 same phenomena were again observed. It should be added that the 

 exposure was not begun vmtil the arc had been struck, and that it 

 burned steadily until the shutter was closed. 



In view of the fact that these lines occur in a nonnal iron arc, 

 it is the writer's conviction tliat the term '' spark " line is misleading. 

 The term "polar'' lines is suggested to distinguish those occui'ring 

 most strongly at the poles of the arc or spark from those occurring 

 most strongly at the centre, for which the term '" median " lines 

 seems suitable. 



The phenomenon does not seem capable of being referred to a 

 "temperature" effect, a conclusion which has been strengthened by 

 Dr. G. A. Heinsalech's notable research upon spectra emitted by 

 flames at different temperatures — he finds, for instance, that the 

 low-temperature Bunsen flame gives a spectrum consisting almost 

 entirely of the enhanced polar lines of iron, and that, as flames of 

 higher temperature are employed, these lines diminish in intensity 

 relatively to the median lines, at the highest temperature many 

 having completely disappeared. 



A list of 202 polar lines in the iron spectrum between XX 2350' 

 and 3,500 has been compiled, the polar lines becoming rarer 



* Fowler. Monthly Notices, Royal A.stroiioniie;il Society, 67, IM, 1907. 

 t Duffiekl. Astrophysical Journal, xxvii., 2C0, 190S. 



